Impact of the economic crisis and contractions within the European long-term care systems

Abstract In 2008, the world underwent one of its worst economic and financial crises, whose consequences are still visible in some countries. This paper aims to analyse the impact of the crisis within the long-term care systems of Germany, England, Sweden and Spain from a comparative perspective. The time period analysed spans from the outset of the crisis in 2008, up to 2017. This article starts off from the thesis of the divergent impact of the economic crisis in these countries and the convergence between the impact of the crisis and long-term care contractions in the most afflicted countries. The outcome highlights the power of economic and financial pressures in order to explain the contractions within the care policies. Equally, it emphasizes the contradictions between the formal development level of the care systems and their practical institutional implementation in the field.


Introduction
Today, challenges and tensions arising from the need to respond to the new social risks (NSR), which are the product of increasingly aging societies and the financial pressures that countries face, are obvious (Starke, 2008;Taylor-Gooby, 2004). Despite this, the severe impact of the 2008 economic crisis has been considerably deleterious for the Welfare State (WS), especially for the Southern European countries (Moreno, 2013). Pierson (2001) has provided an analytical framework to understand how restructuring processes take place in the WS. This very framework has been used to accomplish the main goal of this work. According to Pierson (2001), the cost-containment restructuring process has to do with fiscal pressures and the need to curb public deficit to which the governments are currently subjected; a situation that forces states to implement policies of cutbacks and cost reduction, contracting the WS (retrenchment) in order to alleviate their public deficits. Starke (2008) argues that no established body of literature dealing explicitly with theories of retrenchment yet exists, nor is there any generally accepted definition of WS retrenchment and contraction. From more general WS theories, which were originally elaborated to explain WS expansion, retrenchment has been explained via the neo-functionalist thesis. This argues that since the early 1990s, fiscal constraintspublic deficit and debtas a part of financial problem pressures, are considered important drivers to explain WS contractions. From this perspective, the main purpose of this paper is to explain changes, restructuring processes and contractions of the long-term care systems (LTCS) in Germany, England, Sweden and Spain by means of the impact of the economic crisis in 2008. The term "contractions" is understood here as the reduction of the public longterm care (LTC) protection levels, which would comprise cutbacks and retrenchment in this social policy sector. On the contrary, the expansion of LTCS would occur when there is an increase in the levels of protection. The questions to be answered are: Has the 2008 economic crisis had the same impact on the WS of all the countries studied, and if not, in which did it have a greater impact and why? Would the economic impact of the crisis explain the contractions introduced in the countries' LTCS? The preliminary hypothesis of this study was: "the 2008 economic crisis has provoked contractions within the LTCS of all the countries, mostly in those in which the crisis had a major impact in intensity and duration." As analysed throughout this paper, the first part of this preliminary hypothesis will be refuted, but not the second one. Greater intensity refers to a larger decrease in economic growth in terms of Gross Domestic Product (GDP) as well as larger rates of unemployment, deficit and public debt. Duration refers to the temporal prolongation of the economic crisis in these countries.
This work is framed within the comparative analysis of the LTC European social policies and covers the timespan 2008-2017. The methodological framework used has been cross-national and case study analyses. The case studies correspond to the classic categorization of the Welfare Regimes of Esping-Andersen (1990): Conservative (Germany), Liberal (England) and Social-democratic (Sweden). Spain is also included as a representative case of the Mediterranean WS model (Ferrera, 1996).
The analysis of the LTCS is focused on four dimensions: nature and principles, access of the system, services provision, organization, management and funding. The system's nature and principles refer to the access criteria, which can be: universal, contributory or assistance-based. In countries with a universal nature model, all citizens in need are protected without any restrictions. The contributory model is based on the contributions paid to Social Security. Within the liberal or assistance model, access depends on means-tested income (Esping-Andersen, 1990). The access dimension analyses formal proceedings, from the outset until the end of the processwaiting lists, access criteria, scale assessment and so onwhich enable citizens' access to LTCS. Despite the huge variety of types of provisionpaid and unpaid, work leaves, informal, amongst othersthey have not been included in this study. The service provision dimension in this study mainly assesses the public community and residential services provision. Finally, organization, management and funding analyses the organization and management of the LTCS in terms of global funding, contributions among the different agents, and care labour market conditions.
To explain the impact of the economic crisis, the category of "economic crisis impact" has been used, which includes the indicators of GDP, deficit and public debt rate and unemployment. Likewise, the methodological strategy combines quantitative and qualitative data retrieved from primary and secondary sources. The collection and analyses of the information have been structured in three levels. The first level scrutinizes the quantitative data obtained mainly from the exploitation of the standardized databases of the Organisation for Economic Co-operation and Development (OECD) and the countries' national statistical offices. On the second level, the legislative changes concerning the countries' LTCS are reviewed and analysed following the analysis dimensions. The third level of analysis captures the perception of the social and institutional impact of the crisis from 41 semi-structured interviews conducted with key informants who work in the LTC sector (professionals and managers, scholars and researchers).
The paper is structured in three large sections. In the first, a theoretical approach is deployed for the key concepts related to LTC and the neo-institutionalist theory that serve as a theoretical basis to explain the restructurings from the so-called pressure problems currently affecting these countries: NSR and the 2008 economic crisis. In addition, a review of the most recent literature related to the impact of the crisis on European social policies and LTCS is presented. The second section analyses the impact of the crisis on each of these countries by virtue of macroeconomic indicators and changes within the LTCS in public services, legislation and the social and institutional impact following the above-mentioned four dimensions. The third section presents the concluding remarks of this study.
The long-term care policies in the European framework Initially, the study of care analysed reproductive work connected to activities carried out by women within the domestic sphere. According to Daly and Lewis (2000), there are two major theoretical currents in the study of care: one which focuses on the intimate and private setting, and the second, named "social care," which deals with the study of services associated with the WS and covers: paid and unpaid care; formal and informal, public and private; monetary and in-kind; and contractual and non-contractual provision and which can be defined as: "the activities and relationships involved in meeting the physical and emotional requirements of dependent adults and children, and the normative, economic and social frameworks within which these are assigned and carried out" (p. 284). On the other hand, OECD (2011) has defined LTC as: "The care of people who need support in multiple facets of their lives over a long period of time. Normally, this refers to help with the habitual activities of daily life (ADL) such as bathing, dressing, entering and leaving their homes, which are often left to relatives, friends, and unqualified caregivers" (p. 39). Moreover, despite the fact that a dependency situation can affect any age and stage of life, the elderly are the ones who will require greater support in the future.
With the exception of Northern European countries, LTC policies do not enjoy an established tradition in the European framework. Actually, it was not until the 1990s, as a consequence of the growth of care needs, that LTC was included in the political agenda of most European countries. The WS regimes proposed by Esping-Andersen (1990) confer specific features to European LTCS. In the universal-based models, all citizens have the right to access the system without any restriction and regardless of their income. This is the case of Sweden and recently Spain. The constitution of the Spanish LTCS has a short trajectory. In Spain, Law 39/2006 of the Promotion of Personal Autonomy and Care for People in a Dependent Situation (LAPAD) goes from an assistance-based care system towards a universal-based nature care model based on universal and subjective rights.
The Swedish WS has historically been characterized by high levels of universal access and public services provision (Anttonen, 2002). The concept of care is understood as support provided by society to those particularly vulnerable or in a difficult situation. One of the main principles in this system is to minimize family dependence and make the state responsible for care, which displays a "voluntary nature" (Berggren & Trägårdh, 2010;Kautto, 2010).
The English LTCS has been progressively developing since 1990. Since its inception, the programmes are specifically aimed at elderly people with high levels of dependency that do not have family support or economic resources (Ikegami & Campbell, 2002). The subsidization by the state to the family and the market in the coverage of care needs by virtue of the purchase of private services by families constitutes one of the distinctive features of the English LTCS (Comas-Herrera, Wittenberg, & Pickard, 2011).
Germany is one of the most representative cases of the Conservative WS, considered as "Social Security States" (Palier, 2010). The LTC Insurance Law, Sozialgesetzbuch SGB XI, is based on the subjective right submitted to Social Security contributions and labour. All insured persons, irrespective of age and economic capacity, have the same rights. The principle of family subsidiarity prevails, in which the family is the main party responsible for care (Arntz et al., 2007). This law regulates both the public or statutory LTC and private insurance. Both have recently become mandatory (Schulte, 2009;Schulz, 2010).

The pressure problems of the countries
The new social risks The neo-functionalist theories, which include the "pressure problems" (Schwartz, 2001), have become relevant as explanatory factors of the transformations of the WS. This thesis uses variables such as the economic cycles or the NSR to explain the reforms and defines a new structure of risks or needs arising from the socio-demographic transformations in post-industrial societies (Bonoli, 2006;Taylor-Gooby, 2004). The NSR are the result of changes in the demographic structure stemming from aging and increases in life expectancy. Family changes have also contributed to the transition from a traditional family modelmale breadwinner and female caregivertowards a family dual-earner model, in which both family members work. The rise of the educational level of women and their widespread incorporation into the labour market in the last decades elucidate why many women are neither willing to nor able to, take on the explicit role of carers anymore (Daly, 2012). This situation has crystallized in the so-called crisis of the informal care model or care deficit featured by a progressive reduction of informal support in increasingly aging societies (Daly & Lewis, 2000;Hochschild, 1995). As shown Table 1, from a broader view, there is an estimated decrease in population groups under 64 years of age and a significant increase in the group of people over that age.
The economic crisis of 2008 in Europe: impact and development The economic and financial crisis originated in the United States in 2007 has been considered as one of the worst global economic crises since The Crash of '29 (Castells, 2014). The study of the economic crisis in the WS has been approached from different perspectives and sectors of social policies. Farnsworth and Irving (2012) considered that the crisis affected countries differently and a new international "austerity consensus" has appeared. Vis, Van Kersbergen, and Hylands (2011) analysed the political response to the economic crisis within a set of European Countries, pointing out that there was not a major onslaught against the WS in the immediate wake of the crisis, but also warned about the arrival of more austerity periods to re-establish the balanced budgets lost in the first phases of the crisis. According to these authors, the initial response policies were, in general, more in line with Keynesian intervention and compensation hypothesis rather than a retrenchment cutting back the WS. Consequently, governments rescued the financial sector, introduced measures to stimulate demand, and protected people against the misfortunes provoked by the crisis, which meant that the WS was included in the solution of the crisis. However, after this first stage, owing to the economic crisis, the fiscal consolidation measures served as a springboard and accelerated a progressive process of the European Social Model reforms. Hence, we can distinguish between two phases of the crisis. The first, from 2008 to 2010, which as a consequence of the activation of automatic stabilizers, involved a significant rise in public social expenditure (6% EU27). In this phase, the social protection mechanisms helped to curb the social cost of the crisis and prevent the collapse of consumption and the loss of citizens' purchasing power (Vaughan-Whitehead, 2015).
However, from 2010 on, hefty fiscal consolidation measures to cope with the economic pressures were put in place in several European countries due to the increase of public deficits and debts. These measures were particularly conspicuous in countries where the economic situation was worse, and labour market pressures were greaterthe case of Italy, Portugal, the United Kingdom or Spain (Greve, 2012). Conversely, changes were significantly less severe in countries where the debt crisis was less acute (Vaughan-Whitehead, 2015).
Since 1990, Pavolini and Ranci (2013) and Ranci and Pavolini (2015) have analysed institutional and policy changes in a set of European countries identifying three main problem pressures or drivers: sociodemographic, socio-cultural and institutional and financial pressures. Socio-demographic pressures are connected to NSR. Socio-cultural and institutional pressures explain changes by means of a substantial shift in the political and cultural attitudes towards care provision. Finally, financial pressures may stem from rising social care demands by the growing number of dependent people, the increasing cost of WS programmes, and changes in the financial and economic conditionsdebt and deficit surplus in relation to national GDP, which would lead countries to cost-containment (Pierson, 2001). After the 1990s, according to Pavolini and Ranci (2013), although the problem pressures described above are similar in all countries. These pressures have affected countries differently depending on their specific model of LTC regime. In the universalistic LTC regimeexcept Swedenthe changes have affected the provision of services without altering the fundamental rights of the system. Actually, cutbacks and costcontainment in financing and service provision were introduced to reduce public expenditure. Hence, social care entitlements were not formally reduced, yet a hidden attack within the fundamental rights of the system took place in Sweden, as well as in England's. In England, after the 1990s, local authorities used to be the funders and social care providers, but as a rise on LTC demands and ensuing costs evolved, the shift of funding responsibilities to social assistance evoked a widespread marketization in care services. Over the last years, the situation has turned worse since the implementation of organizational measures, targeted at cost-containment and more efficiency in care provision, resulting in a complete transfer from the public sector to third sector charities and for-profit entities. In countries with a Residual LTC model, such as Germany and Spain, the high pressures caused by the rising care demands, which were matched by social services and managed at a local level, were the main reason why German set up a new LTCS in 1995. However, in Spain, the principal hurdle was to find a way to finance a national LTC system. In the study of the Spanish case, the deep contraction in the LTCS since 2010 and its successive reforms generated a significant deterioration in the care and well-being of dependent people (Autor, 2013;Autor, 2015).
The considerable impact of the economic crisis has evoked a vast number of comparative studies in the realm of European social policies. Although these studies provide a long-range scope of the development and reforms that have taken place, there is little development of comparative studies analysing in-depth changes and impact of the economic crisis on LTC policies. For this very reason, this study strives to analyse the changes in the countries' LTCS after the crisis.
Analysis of the impact of the economic crisis in Germany, the United Kingdom, Sweden and Spain In this chapter, the category "impact of the economic crisis" which touches on the following variables: intensity (the GDP, the public debt and deficit and unemployment) and duration (crisis lifespan) are scrutinized. The analyses focus on a pre-crisis (2000-208) and a post-crisis period (2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016) in order to underline the previous economic situation within the countries. Factors, recovery and measures implemented by governments are also discussed.

Gross Domestic Product
Germany is the leading economic power in Europe. In the pre-crisis phase (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008), the German GDP rose from 2,242,493 million U.S. dollars (MUSD) to 3,122,547 MUSD and was reduced to 3,036,964 MUSD between 2008 and 2009. The principal factor explaining this contraction is the reduction of German global trade demand in the export sector (Brenke et al., 2013). As in 2009, the German GDP reached 3,210,822 MUSD and escalated to 4,028,264 MUSD in 2016. The factors leading to Germany's early recovery included the flexibility of its labour market, the bank rescue policies, the sustainability of the economy and its trade surplus (Poli, 2015). In 2008, the German Federal Government launched two support and economic stimulus plans for the financial sector, rescuing the main banks hit by the crisis (Dauderstädt, 2013;Hüfner, 2010). Other measures were previous labour market reforms such as Hartz Reform, which would have allowed workers to keep their jobs at the expense of high levels of flexibility and job insecurity, but it enabled Germany to emerge from the crisis with one of the lowest unemployment rates of the European Union (EU) (Méndez, 2013).
From 2000 to 2008, the UK GDP experienced a steady economic growth from 1,533,016 MUSD to 2,229,260 MUSD. This growth was interrupted in 2008 by the arrival of the global economic and financial crisis. Between 2008 and 2009, the GDP suffered a reduction of 86,771 MUSD. The high extension of its financial sector and the real estate "boom," in conjunction with the high household debt, place this country in a position of special vulnerability in the face of the crisis (Giudice, Kuenzel, & Springbett, 2012). Between 2008 and 2009, some of the main banks in the country went bust, which meant that huge amounts of public money were used to stabilize the financial sector (Heinrich, 2015).
Sweden: Even though the 2008 economic crisis was one of the worst recessions for Swedeneven worse than the ones in 1931 and 1990experts have emphasized the strong resilience shown by this country in the face of the crisis, despite its repercussions in the financial and housing sector (Hassler, 2010;Irwin, 2011). Between 2000 and 2008 the Swedish economy displayed a substantial increase with its GDP going from 259,680 MUSD to 385,876 MUSD. Between 2008 and 2009, the Swedish GDP slumped to 369,089 MUSD. Some of the country's major banks were nationalized after their bankruptcy. With the creation of the Bank Support Authority, the State ensured the supervision and recapitalization of the financial system, assumed the debts and guaranteed all the nation's bank deposits -114 banksin exchange for the purchase of shares that would eventually allow the recovery of the money invested (Dougherty, 2008). Between 2009 and 2010, there was an increase in the GDP that reached 390,766 MUSD.
Spain: During 2000-2008, the Spanish GDP went from 872,912 MUSD to 1,537,767 MUSD. In this expansive phase, spectacular growth of the housing sector took place (Carreras & Tafunell, 2010). From 2008, Spain faced a double crisis: the global financial crisis and the one evoked by the outbreak of the "real estate bubble" as a consequence of the urban excesses of the previous stage. Between 2008 and 2009, the Spanish GDP fell from 1,537,767 MUSD to 1,506,253 MUSD and continued to decrease during 2009-2010. In 2010 and 2011, there was a slight recovery, but in 2012, there was a recession againsee Table 2. During 2008, 50% of real estate agencies closed their doors in Spain, which led to the bankruptcy of a large number of construction companies and the loss of employment for a vast number of workers. The loss of work and high household debt led to a rise in payment default. Since mortgage payments could not be met, many families lost their homes (Espiño & González, 2008).

Public debt and deficit
Prior to 2008, all countries maintained public debt of less than 70% of GDP, but from then onwards, there was an increase in the public debt - Figure 1 in all countries except in Sweden, and to a lesser extent in Germany. The irruption of the economic crisis led to a substantial rise of the public debt and deficit due to the growth in public spending directed at the economic reactivation and the rescue and stabilization of the financial sector.
The countries which best handled this situation were Sweden and Germany. The latter reformed the Constitution in 2009 while introducing budget balance and limiting its debt capacity. This fact facilitated the reduction of the German debt volume to a virtual zero by 2012. Other aspects to bear in mind in the rapid German recovery were the non-impact of the housing crisis and the maintenance of its low levels of  unemployment in the recessive phase. The Swedish government had already implemented fiscal reforms during the 1990s crisis that forced them to keep a budgetary surplus target of 1% and to limit indebtedness as a measure of protection against the crisis. Additionally, the strategy followed by Sweden to face the bank rescuethe purchase of shareswas surprisingly effective, and it permitted them to exit the crisis in a reinforced manner. On the other hand, Spain and the United Kingdom were the countries most affected by the crisis. The exposure of their financial and real estate sectors turned into a spectacular rise in public debt and deficit from 2008 onwards. The insolvency of the construction sector and the bursting of the housing bubble had a substantial impact on their economies and triggered their unemployment rates, especially in Spain. The high exposure and breadth of the United Kingdom financial sector placed the United Kingdom in a situation of special vulnerability. In Spain, the payment of the banking debt made the deficit surpass 11% in 2009. This increase in public deficit primed the conservative government to put in place a severe plan of austerity and budgetary control imposed by the EUsee Figure 2. Spain is, without any doubt, the country most strongly hit by unemployment with a rate of 26% between 2007 and 2013, which remained at 20% throughout 2016 - Figure 3. This situation may be related to the specificity of the Spanish labour market, its historical weakness, the scarce diversification of its productive structure and the considerable weight of the construction sector in the economy. So, after a generalized economic expansion, there was an abrupt decline of GDP in 2008 and throughout all the states, marking the onset of the economic recession cycle and endingwith the exception of Spainin the period 2009-2010. The crisis had an uneven impact on the states in intensity and/or duration. Spain, in conjunction with England, represented the two countries in which the economic crisis had a devastating impact. In these two countries, the intensity of the crisis can clearly be seen via the ensuing high levels of deficit and public debt. The unemployment rate in Spain remained considerably high, the economic recovery lasted longer and was much slower than in the rest of the studied states. Among the four countries, Spain and the United Kingdom had the greatest exposure to the housing and financial sector. Conversely, the crisis had a minor impact within countries with a powerful industrial economy, wider export sector, and superior investment in research and development, as in Germany and Sweden.

Analysing changes in LTC: Services, legislation and actors
This chapter studies the leading changes within LTCS in the countries. The sub-sections correspond to the three levels of analysis of the research: the first level of analysis provides an insight of the changes in  the statistical indicators: total number of workers and carers, number of public places in residences and number of beneficiaries in residential and community/homecare services. The second, of a theoretical and basically qualitative nature, takes a closer look into the legislative and normative changes considering the outset of the dimensions in the methodological section from 2008 to 2017. The third level, also of a qualitative nature, offers the vision of the social and institutional impact of the crisis in LTC.

Public services provision (fist level of analysis)
Next, Table 3 shows the evolutions in the total number of places in residences in Germany, the United Kingdom and Sweden.
When exploring the changes and evolution in the provision of LTC public services in Germany, and according to the latest available data -  in residential services remained flat throughout this period. The growth in the number of residential services beneficiaries during the 2000-2007 period was 20%, and between 2008 and 2015, it was 13.3%; 6.7% less than in the previous period. However, community and homecare services between 2000 and 2007 grew by 9.1% and between 2008 and 2015 by 39.7%; a substantial increase compared to the first period, which shows a clear communitarian-oriented German provision model. Since the OECD fails to disaggregate data on formal workers and beds in residential facilities in England, the UK database has been used to analyse overall trends in these two indicators, and the England Health & Social Care Information Centre was used to examine residential and communityhomecare beneficiaries. The first indicator analysed refers only to non-professional carers. In general terms, between 2003 and 2007 (the data was not available before 2003), the total number of formal carers was 31.9% less, and between 2008 and 2016after a considerable increase as a result of the change in the accounting systemthe total reduction was 22.6%. The analysis of these data might indicate that the contraction of formal carers in the United Kingdom responds to a long-term process rather than a result of the crisis. In the period 2003-2016, there was a total reduction of 0.2% in the number of residential bedsbetween 2003 and 2007, the reduction was 1.4% while from 2008 to 2016, the places increased by 1.2%. Additionally, it can be stated that in 2010, the number of places fell by 3.7%. Thus, despite the rise in 2011probably as a consequence of the changes in accountability of servicesthe tendency towards the reduction in public residential beds in the United Kingdom is blunt. The number of both residential and community-homecare beneficiaries between 2008 and 2014 (only available data) refers exclusively to England. In this case, a predominance of homecare provision is also appreciated. In 2008, the total number of beneficiaries of services was 1.8 million, of whom 86% were eligible for homecare services, and 14% were attended in residences. Between 2008 and 2014, the total figure of citizens receiving homecare services was reduced by 480,000 (31% less), and the number of people eligible for residential services also fell by 29,000 (11.6%) in the same period, which shows an evident overall contraction in the number of beneficiaries in the LTC services.
The total number of workers in the Swedish LTC sector includes both professional and nonprofessional workers. Still, the overall balance of workers in the LTC sector between 2000 and 2015 is positive, given that it increased in absolute numbers from 211,764 to 237,583; a total increase of 12.2% (by 5.7% between 2000 and 2007, and 6.15% between 2007 and 2015). The number of workers in the period 2006-2009 was reduced by 3. 4% (0.2% between 2006 and 2007, 0.8% between 2007 and 2008 and 2.4% between 2008 and 2009). (see Figure 4).
Between 2000 and 2015, the number of places in residential services in Sweden went down from 150,939 to 125,386: a 17% reduction. The slump began in 2002 and, although there was a small increase in 2008 (0.5%), it continued progressively falling until 2014. In the first period 2000-2007, there is a 10.8% reduction in the number of residential places, and in the second period between 2008-2015, the reduction was 6.8% in total. In Sweden, from 2000 to 2015, the total number of beneficiaries of residential services decreased by 28,052 peoplefrom 141,850 to 113,978 beneficiariesa decrease of 20%. By contrast, in the same period, there is a huge increase in the number of home-services beneficiaries which went up from 193,189 to 328,858 as Figure 5 shows, there is a clear trend change in the provision of services from 2001 onwards; home services are progressively increasing while residential services show a continuous reduction. The most important aspect is the spectacular increase in the number of community and home services from 2005 onwards.
In the Spanish case, there are no data available until 2009 as the LAPAD was not put in place until 2007. Moreover, figures for places in residences have not been included in this section either, given the lack of reliability of data according to the OECD. During 2009-2016, the total number of workers in the LTC sector increased by 21.7%. However, after a closer look, it can be seen that in 2010 and 2011, there was a fall of 0.5% and 0.6%, respectively in the number of workers. The total number of beneficiaries of residential services in Spain in 2009 and 2016 increased from 117,346 to 208,108, ie. by 77.3%. In addition, the figures of beneficiaries of homecare services increased by 126.8% from 379,564 to 860,859. As far as the inter-annual variation is concerned, in the first years of LAPAD implementation, the number of beneficiaries of residential services increased by 40%, but since 2011 it remained nearly frozen with a small increase in 2015. Similarly, by the end of 2011, the increase in homecare services had risen to almost 50%, yet at this point, a regression started, reaching À2% in 2013. Figure 6 clearly shows the expansion of the Spanish LTCS. Ever since, we can discern a predominance of community and homecare services over residential services.
To sum up, the data show that in Germany, an overall expansion of the LTCS has occurred via reinforcing its traditional principles and nature of community and family life. This expansion has mainly taken place by means of the growth in the figures of beneficiaries of community and homecare services. Conversely, the number of beneficiaries in residential places has contracted in relation to the previous period (2001)(2002)(2003)(2004)(2005)(2006)(2007). This fact will advocate a model with a clear communitarian care provision approach, and it reinforces the idea of the re-familiarization of the system, as a result of this increase in the provision of care in the family environment.
In the United Kingdom and England, it can be stated that there has been a significant overall contraction in all the indicators analysed. The contraction in the number of workers in the formal LTC sector in the precrisis, as well as post-crisis periods, has been significant; ie. that cutbacks in the care sector had already started at the onset of the crisis. In England, a predominance of homecare provision is also appreciated; in 2008, 86% of beneficiaries were receiving homecare services, and 14% were attended in residences. The reduction in the total figures of beneficiaries receiving homecare (31% less) and residential care services (11.6% less) during the post-crisis period indicate an overall reduction in the system.  In Sweden, although the total number of workers within the LTCS increased slightly, a reduction of 3.4% between 2006 and 2009 is visible. As can be seen, the number of places in residential services during 2000-2015 fell by 17%, and also the number of beneficiaries in the overall period by 20%. Meanwhile, the number of community and homecare service beneficiaries increased by 70% in the overall periodie. by 32.2% between 2008 and 2015. In this case, the tendency suggests a slight reduction in workers during the crisis period, a clear retrenchment and replacement of residential care services towards community and homecare services which took off after 2008.
The Spanish LTCS shall be explored under a double process of implementation-expansion and reform-contraction of the system. Its launch fully coincided with the outbreak of the economic crisis, whose impact was to be felt from 2010 onwards with a series of legislative reforms which would radically alter and contract the systemas analysed in the next sub-section. Indicators show a generalized expansion of the system, especially in community and homecare services since the beginning of the system. It is therefore worth mentioning that the Spanish LTC system has been the one that has expanded the most, resulting from the implementation of LAPAD since 2007.

Legislative changes (second level of analysis)
In this second subsection, legislative changes undertaken by the countries, following the main analysis dimensions, are examined.
The legislative reforms applied to the German LTCS during this period were Laws I and II in order to strengthen the LTCS -Pflegestärkungsgesetze PSG-I-II. There are no significant changes in the contributory nature (dimension 1) of the system because its expansion and reinforcement have been accomplished by increasing the workers' contributions. The inspiring principles also continued to establish the subsidiary of care within the family and community realm, which includes proximity in the provision of services, and the promotion of prevention and autonomy. Changes in dimension 2 (access of the right) have had as a primary objective to redefine the profile of the users and the attention needs. Therefore, a new scale (from 3 to 5 degrees) was introduced in order to extend coverage to people with cognitive disorders and other mental disabilities, who were excluded from the previous scale. Dimension 3 (service provision) has been one of the most reinforced. A general increase in the provision of economic benefits and services occurred, in all the degrees of dependence. However, the trend was towards strengthening family support at home or in the community. In the fourth dimension (organization, management and financing), the most crucial aspect in Germany would be the creation of a reserve fund to meet future care needs arising from the aging population.
In England, the legislative framework and reforms, introduced in the studied period, correspond to the Care Act of 2014 and the Welfare Reform of 2012. In relation to the analysis of dimension 1 (nature and principles of the system), the new Care Act places particular emphasis on prevention and promotion of well-being principles. The principle of prevention is reinforced by diminishing the level of dependence, so that people with lower levels of care needs can access earlier the system. In dimension 2 (access of the rights), this law introduced an important change within the scale of assessment compared to the previous scale (Fair Access to Care Eligibility Bands 2002), while extending social care to people with lower levels of dependence (or care needs). Reversely, with the reform of the 2012 Welfare Law, the criteria and conditions for accessing financial services for people with disabilities were restricted. The legal expansion of access, provided by the Care Act, also affects caregivers who from then on have the right to be assessed and oriented about resources to meet their needs on a universal base and irrespectively of their income level. As far as the services provision is concerned (dimension 3), the Care Act compels municipalities to apply the principle of free choice, which is supposed to respect the will of the "consumers" by bolstering the consumer-customer concept.
Finally, related to dimension 4 (organization, arrangement and financing), the Care Community Commission created by the Care Act was conceived to supervise the work carried out by municipal social workers to guarantee the system's economic efficiency. In the same vein, in 2012 the Welfare Reform Act put into place a penalty system to control the economic benefits for people with disabilities. In the realm of funding, the Care Act enlarged the threshold for accessing to public services by raising the income limit to access from 23,500 to 72,000. Moreover, the law established a maximum amount of the cost of the services to prevent that people with severe dependence were excluded from the system, because they fail to afford to pay for the costly care services. However, all these economic improvements did not come into force due to the austerity measures, the cost-containment and cuts in the services funding introduced by the government in the wake of the economic crisis.
In Sweden, the LTC system has undergone minor legislative reforms since 2008. The most significant reforms introduced between 2008 and 2017 touched on dimensions 2 and 4. The changes in dimension 4 mainly affect the quality, coordination and general organization of the provision system. As far as provisioning is concerned, the implementation of a free choice of provider means the definitive establishment of a private management and public financing provision model.
The Spanish LTCS, concerning dimension 1, after the enactment of the LTC law in 2006, shifts from a system of an assistance-based nature to a universal-based system. Even with this normative deployment, the reforms introduced from 2010 onwards prompted rights and principles, which were envisaged by this law, to not become effective. In 2008-2010, the modifications are still oriented towards the development of the LAPAD and the expansion of the system. Reversely, from 2010 onwards, the legislative changes basically affected dimensions 2 and 3. These changes were geared towards an overall LTCS cost-containment and retrenchment, restricting the terms of access and reducing the number of services. Moreover, in dimension 4, the co-payment of the services increases, and so does the economic, personal income for calculating the services, according to the regions.

Social and institutional impact (third level of analysis)
The people interviewed in Germany have a generally positive perception of the improvements introduced, even if there has been no change in the nature and principles of the system (dimension 1). In particular, they appreciated the expansion of access (dimension 2) as a result of the modifications within the assessment scale, which has enabled people with cognitive disorders and psychological impairments access to the system. In dimension 3 (service provision), the perception is that there has been an expansion in economic benefits accompanied by greater control of spending or even cutbacks in services for people with disabilities. There has also been a certain containment in residential services entailing a respectable increase in waiting lists in recent years and, consequently, clear prioritization of community care over residential. As for the changes discerned in dimension 4 (organization, management and financing), there is a general perception of a shortfall of specialized care professionals and the emergence of an informal deregulated care market.
In England, the perception in dimension 1 is that the system is basically oriented towards covering the basic needs of the poorest. In dimension 2, the professionals sense a clear trend towards restricting access to residential services, resulting in a rise in waiting lists, and the prioritization of attention in the community environment which could induce a desire for cost-containment expenditure. The main changes detected in dimension 3 have been the curbs in public service provision and the prioritization of the least expensive services. At the same time, a reduction in the supply of private services is also visible in some territories, as a result of the disappearance of non-profit entities and private agencies dedicated to care. All these factors bring us closer to a clear trend towards cost-containment and a certain crisis in the provision of care. Finally, in dimension 4, the consequences of the cuts in the financing of municipal services are once again evident; this is reflected within the reductions in the hiring of staff, and the increase within the workload of professionals who are unable to carry out the high workload in assessing and attending to people in need of care. All these factors clearly show an overall reduction in English LTCS.
In Sweden and dimension 1, the interviewees perceive that despite the strengthening of the "aging at home" principle and the promotion of "free choice of provider," the Swedish LTCS still upholds its universal-based nature and fundamental principles. Concerning dimension 2, the access extension to family caregiverswho from 2009 onwards are eligible for supportwould be the outcome of communitarian policies linked to the restriction of criteria for accessing residential services, which would have prompted more people with severe dependencies to stay at home being cared for by family members. With regards to dimension 3, front-line social workers made clear that the restrictions issued within the assessments to access the services had been used for cost-containment in the public services and to encourage the purchase of private services. The changes mentioned in relation to dimension 4 have been the definitive introduction of the New Public Management criteria in the public services and the massive establishment of private provision management in the services. These modifications in provision have configured a new private care market monopolized by large corporations which, according to experts, has led to a significant loss in the quality of services, precariousness of the care work sector and a crisis of qualified professionals.
In Spain, the organizational and financial problems meant no real coverage of the needs of dependents, and the principles (dimension 1) proclaimed by this law were not made effective. With regards to dimension 2, those interviewed highlighted the strategies used by the Administration for costcontainment by delaying and restricting access as much as possible in order to save money. Regarding dimension 3, the professionals claimed that although the contractions did not begin until 2010, the endemic provisioning of services had been felt since the outset of the implementation of the LAPAD. From 2010 to 2012, changes in this dimension were aimed at reducing costs and generally contracting the system. As for the changes in dimension 4, it should be clarified that the financial problems and the progressive withdrawal of central government funding from 2010 onwards would have led to significant disparities and territorial imbalances in the implementation of the LTCS.

Conclusions
First, it can be stated that the economic recession of 2008 has had an unequal impact on the countries, both in intensity and duration. Spain and England represent the countries in which the crisis has had the greatest repercussions: the intensity of the crisis is evident from the sheer levels of deficit and public debt reached. Moreover, in the case of Spain, unemployment rates remained high, even long after the outbreak of the crisis, and they continued to do so until 2016. Second, the crisis had a smaller impact on those countries with a larger productive sector, high levels of exports and production linked to great investments in research and development -Sweden and Germany. Third, the crisis had greater repercussions on those countries whose main economic sectors were tightly linked to the crisis, a bigger financial and real estate market -England and Spain. Another aspect to bear in mind is that some of the countries have some protective factors against the crisis. These are Germany and Sweden whose measures to control the deficit and the public debt or the labour reformsintroduced by Germany some years earlierwould have enabled tighter control of unemployment during the crisis.
These first conclusions determine two clusters of countries: on the one hand, England and Spain whose LTCS have endured major contractions and the crisis had a more dramatic impact. On the other hand, Germany and Sweden in which the economic crisis had a minor impact, and their LTCS did not contract or contracted lesssee an outline of contraction-expansion of LTCS according to the analytical levels in Table 4.
The second conclusion: as a singular and contradictory aspect (except in the English case where the amendment of the Welfare Reform Law produced a contraction within the benefits), there has been a widespread expansion of legal social care rights in all countries. However, in real terms, only the German LTCS has benefited from a real and general expansion. At the other extreme, we can find the English case whose analysis shows a significant and overall contraction of the system in spite of the improvements stated by the Care Act. In Sweden, we do observe an expansion in the provision of community services, a contraction in the number of workers, and a general contraction of the residential services. Finally, in Spain, there are two contrasting dynamics of expansion-contraction of the system: the first phase coincides with the implementation stage of the system between 2008 and 2010, whereas the second phase occurs from 2010 onwards with the implementation of the first austerity measures and control of the public deficit. The definitive displacement of the residential services for the sake of community services is also apparent.
Lastly, and responding to the questions raised in the introductory section of this research, it can be stated that the crisis had a divergent impact on the LTCS of the studied countries: those enduring a greater impact are undoubtedly those countries in which the LTCS have contracted the most. Therefore, it can be inferred thatwithout detriment to the explanatory value of other variablesthe impact of the economic crisis of 2008, namely, financial and economic pressures introduced in the theoretical framework, contribute to explain the changes geared towards cost-containment and retrenchment of LTCS.
Notes on Contributors. Dr. Esther Albesa is a professor of Social Services and a collaborating member of the research group AINSR (Analysis of Inequalities and New Social Risks) at University of Barcelona. Her current research interests comprise social policies and reforms in European long-term care systems. She is a member of the executive board in the Catalan Sociology Association. Her work has been published in the Spanish Journal of Social Services, Catalan Journal of Social Work, and in the Journal of Spanish Network of Social Policies, among others. She has been working for the regional and local administration for since 2002. (3) Perception of the interviewees

Contraction of residential services
General contraction of the system

Contraction of residential services
General contraction of the system > 2010 Source: prepared by the author.