Commission Staff Working Document - Demography Report 2008:Meeting Social Needs in an Ageing Society

3.1. Older people in employment ...................................................................................... 94 3.2. Senior citizens' involvement in unpaid work ........................................................... 108

3.3. Responding to the needs of an ageing population: policy implications ................... 122

3.3.1. Policies to promote active participation in employment and society ....................... 122

3.3.2. Policies to preserve the autonomy of older people................................................... 123 4. Tackling the Challenges of Demographic Change: Update ..................................... 127

4.1. Improving the conditions for Europe's demographic renewal.................................. 128

4.2. More employment .................................................................................................... 133

4.3. Higher productivity .................................................................................................. 135 4.4. Receiving and integrating migrants.......................................................................... 141

4.5. Sound public finances............................................................................................... 147 Annexes .................................................................................................................................. 150

1.

Country Annex ......................................................................................................... 153
2.

Statistical annex........................................................................................................ 212

S UMMARY

Member States can tackle the challenges of demographic change by acting in five key areas. In October 2006, the Commission presented its views on the demographic challenges the EU faces and on opportunities for tackling them in the communication "The demographic future of Europe --from challenge to opportunity" 1

. The communication expressed confidence

in Europe's ability to adapt to demographic change and notably population ageing, but also stressed the need to act in five key areas:

Better support for families; Promoting employment;

Reforms to raise productivity and economic performance; Immigration and integration of migrants; Sustainable public finances.

The Communication also announced that every two years, the Commission would hold a European Forum on Demography. The first Forum took place on 30-31 October 2006, the second on 24-25 November 2008. The fora are an occasion for taking stock of the latest demographic developments and reviewing how policies are responding to demographic change.

Member States face different sets of opportunities. This report presents comparative data to national policy makers. The purpose of this second report is to provide the latest facts and figures that are needed for an informed debate with the stakeholders taking part in the Forum and with the group of government experts on demography, which was involved in the conception of the present report.

As far as possible, data are provided for each Member State, allowing policy makers and stakeholders to compare their own country's situation with that of others, to understand the specificity of their country and, possibly, to identify countries that provide interesting experiences from which others can draw lessons. In so doing, the report responds to the request from Member States who want to learn from the variety of national experience across the European Union. The report focuses in particular on two issues that have received much attention following the adoption of the communication on Europe's demographic future: the modernisation of family policies

and opportunities for enhancing the

contribution of older people to the economy and society .

1 COM(2006) 571, adopted on 12 October 2006. 2

See the Communication from the Commission Promoting solidarity between the generations, COM(2007) 244. 3 See Council Resolution of February 2007, DOC 6216/1/07.

Population pyramids' show how births, deaths and migration shape the structure of a population... Chapter 1 looks at the latest data on the determinants of Europe's demographics, namely, births, deaths and migration, and presents the latest population projections from Eurostat. These three factors shape what are known as 'population pyramids', which show the structure of a population by sex and by age. Describing such a chart as a 'pyramid' is quite inappropriate, as far as the EU or other developed countries or regions are concerned. The pyramid shape is characteristic of countries with high birth rates, and thus rapid and unsustainable population growth, or countries with a high mortality at all ages, or a combination of both. Europe has been very successful in tackling both of these demographic challenges. ... but in advanced societies with stable populations the pyramid turns into a pillar. A desirable shape for the 'population pyramid' chart is more like a pillar, which would result from a steady renewal of a stable population, with almost all new-born children surviving into old age. The EU is coming close to such a shape, except for the bulge of the baby-boom that started in the 1950s and reached its peak 20 years later. Since then, the number of births per woman has declined significantly: women born in 1935 had on average (EU-25) 2.37 children, those born in 1945 2.11 and those born in 1955 1.94. Women born in 1965 have had 1.77 children, but their total cohort fertility rate can only be established once they reach the end of their reproductive period. Population structure in 2008 and 2060

EU 27

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) Source: Eurostat, EUROPOP2008 convergence scenario.

he EU fertility rate is today estimated at 1.5 children... The fertility rates of younger cohorts can only be estimated. This is done using the total period fertility rate indicator, which is based on childbearing probabilities currently observed for women of different ages, and derives from them the average final number of children women would have if they realised these probabilities during their lives. This indicator now stands at just over 1.5 children per woman, showing a slight progression between 2000-2004.

...but postponement of births could result in current fertility being underestimated. The problem with the total period fertility indicator is that it is affected by changes in the timing of births. The mean age of women at the birth of their first child rose from 24.4 years in 1960, to 24.6 years in 1980 and 27.5 years in 2003 (EU-25). As a result, the probabilities of having a child are reduced at younger ages, which are captured by total period fertility during the shift towards a higher age. By contrast, the increase in the probability of giving birth at an older age will only materialise later, once the postponement process stops and a shift to a higher average age is completed. The total period fertility rate indicator will, as a result, underestimate the number of children women will have. Demographers have tried to correct this bias, and one such correction lead to the conclusion that actual fertility rates, adjusted for this 'tempo effect', could be almost 0.2 children per women higher than the unadjusted total period fertility rate. This is still below the replacement rate of 2.1, but would make a major difference for Europe's long-term development in terms of age structure and population size. Europeans gained 2.5 extra life years per decade over the past 40 years; future gains will have to result from reduced mortality in old age. The second factor that determines the size of Europe's population is the number of deaths. This depends, on the one hand, on the size of cohorts reaching the end of their life span and, on the other, on mortality rates, which, in turn are used to estimate life expectancy. In 2004, the EU-27 life expectancy was 81.5 years for women and 75.2 years for men. During each of the preceding four decades, around 2.5 extra life years were gained by reducing mortality, due to progress in fighting respiratory diseases and cancer in the 1970s, and cardio-vascular diseases in more recent years. Today, there is little scope for further gains in life expectancy by reducing premature mortality (up to the age of 60) in most Member States. The vast majority of new-born children can expect to live to this age. Any further improvements in life expectancy will have to come from better health in old age. Life expectancy is still low for men in many Central and Eastern European countries and throughout the EU, people of lower socioeconomic status die younger. As far as life expectancy is concerned, there is, however, still a significant East-West divide in the EU, a divide that particularly concerns men, who can expect to live only to around 65-70 years in eight of the Central and East European Member States, compared to an EU-15 average of more than 76 years. In these countries, mortality among middle-aged men remains high, and this is the main factor behind the large East-West gap in life expectancy. Another important divide regarding life expectancy is found within each society: a higher socio-economic status in a society is associated with significantly lower mortality and hence higher life expectancy. These health inequalities have been identified as a major challenge to be pursued in the follow-up of the Renewed Social Agenda 4

4 The Communication from the Commission of 2 July 2008 "Renewed social agenda: Opportunities, access and solidarity in 21st century Europe" (COM(2008) 412) announced that the Commission will issue a communication on health inequalities during 2009, building on work under the Open Method of Coordination on social protection and social inclusion.

Since 2002, net migration has been historically high at 1.6 2 million people per year. The third determinant of population structure is migration. EU-27 has attracted year on year around half a million migrants more than it lost over the previous 20 years. Since 2002, however, net migration into the EU has roughly tripled to between 1.6 and 2 million people per year. Only three countries, Spain, Italy and the UK, have received around three quarters of net migration into the EU over the six years up to 2007. As a result of this immigration, around 4% of EU residents were non-EU citizens at the beginning of 2007, whereas 2.1% of EU citizens were living in a country other than their country of origin. The number of foreign residents depends, however, not only on migration, but also on the rate at which migrants acquire the citizenship of their host country. In 2006, 670 000 third-country nationals became citizens of an EU Member States, about the same number as in the US.

The latest Eurostat population projections expect continued population growth up to 2060... Assumptions about future fertility, mortality and migration are at the basis of projections of the future size and structure of the population developments. The report presents the results of the latest round of Eurostat population projections, and compares them to the previous round in 2004. Whereas the previous projection round concluded that the population of EU-27 was likely to decline by 16 million people by the year 2050, the latest projections expect an increase by 10 million people by the year 2060. Thus, the population of EU-27 would rise from 495 to almost 506 million people. The difference is mainly due to the higher migration assumption, but more optimistic fertility and life expectancy assumptions for the latest projection also contribute to the large difference between the two rounds. There are also striking differences in the results of the two projection rounds for individual countries.

...contrary to projections made four years ago. However, rapid population ageing occurs under both projections. These differences between the two rounds of projections underline the importance of interpreting such results with caution. Nevertheless, one development is certain, namely the imminent retirement of the baby boom cohorts which will shift the balance between people of working age and retirees. Both projection rounds also yield very similar results as far as the long-term evolution of the demographic old-age dependency ratio (people aged 65+ in relation to people aged 15-64) is concerned: in 2004, a ratio of 0.53 was expected for 2050 (EU-25), and the latest projection expects a ratio of 0.50 in 2050, rising to 0.53 by 2060 (EU-27). This is to be compared to today's old-age dependency ratio, which stands at 0.25, meaning that, for every person aged 65 or over, there are four people of working age (15-64). In 2050, there will be only two people of working age for every person aged 65+. Europe is not alone in experiencing such ageing: it is a common trend across the globe and could occur at an even faster rate in emerging economies.

The reality of family life has been transformed: fewer and later marriages, more divorces, unmarried cohabitation and single parent households. Chapter 2 describes the changing family and household patterns in the EU that need to be taken into account in the modernisation of social and, in particular, family policies. The social reality of family life has changed profoundly over recent decades. People are less likely to enter into a first marriage, and, in 2003, did so about two years later than in 1990: the average age at first marriage rose from 24.8 years to 27.4 years for women and from 27.5 to 29.8 years for men. A significant number of marriages are between partners of different nationalities: between 12% and 15% in Germany and France, around 20% in Belgium and Austria, between 25% and 30% in Estonia, Luxembourg and Cyprus. Divorce rates increased since the 1970s, more than doubling in some countries. About 20% of all marriages involve divorced persons entering into a second marriage. Unmarried cohabitation has become common place, and a large proportion of children are born outside marriage: in most Member States between 25% and 50% of all children. In spite of this 'de-institutionalisation' of family life, most children still live in couple households, married or cohabiting, which represented over 80% of households with children in 2001. Single-parent households, most of them headed by mothers, accounted for 14% of households with children. Women are much more integrated into the labour market, but continue to shoulder most of the family responsibilities. Another major trend affecting family life in the EU is increased female labour force participation. The gap in employment rates between men and women aged 25-49 halved between 1990 and 2005 from 32 percentage points to 16. However, this labour market integration of women often takes the form of part-time employment. Marital status, childbearing and childrearing are no longer seen as an insurmountable obstacle for female employment. Indeed, countries with high levels of female employment also have higher fertility rates, suggesting that achieving a satisfactory work-life balance for women may be a key to both higher female employment and a relatively high fertility rates. Average household size decreased from 3.3 in 1960 to 2.4 today, notably as a result of the rising number of singleperson households. There have also been significant changes affecting the composition of households. The average household size in EU-25 declined from 3.3 persons in 1960 to 2.4 in 2003, implying a much faster growth in the number of households than in the population size. In 2005, 27.7% of all households were single-person households, almost the same proportion as family households (two or more adults with dependent children). A large number of single person households belong to people over the age of 80. Nearly 15 million people over 80 live in a private household (as opposed to an institution), and about half of them live alone. The increasing number of older people will further increase the number single-person households. Due to population ageing, the number of one- and two-person households can be expected to increase considerably. As the large cohorts of the baby-boom reach retirement age, more and more people belonging to these cohorts will be living in smaller households. In most European countries the number of these small households is therefore likely to increase by at least 50%, and in several countries the number

could even double between 2001 and 2050.

Policies need to adapt to the changing reality of family life; in particular, singleparent families are at a high risk of poverty. These changing family and household patterns need to be taken into account in the modernisation of social, and in particular, family policies. The need for action is exemplified by the high risk of poverty to which single-parent families are exposed: about one third of people living in single-parent families are at risk of poverty, compared to 16% of the entire population. Large families (two-adult households with three or more children), which can be considered as a more traditional target group of family policies, also face higher poverty risks at 24%, but less so than single-parent families. A quarter of women living in singleperson households are also at risk of poverty.

Cash benefits dominate social protection spending for families and children, but childcare services could play a major role in protecting families against poverty, allowing them to have a second income. A key question is how best to support families. The financial situation of families and the risk of poverty they are exposed to depends on the combination of incomes that parents can earn and the benefits they receive. Benefits may go a long way towards covering the costs of children, but would have to be very high to replace a second income in a two-adult household. Indeed, only about 7% of households with dependent children were at risk of poverty in 2005 when both adult household members were in employment, compared to 16% of people at risk of poverty in the population as a whole. The availability of affordable childcare can therefore have a major impact on the financial situation of families. Social protection spending on families and children 2.1% of EU-27 GDP in 2005 remains, however, strongly geared towards cash benefits: three quarters of this amount is used for cash benefits such as child allowances and one quarter for services. There are major differences between countries in the split between benefits and services in kind, and also in the use of formal childcare, particularly for children below the age of three.

Countries with high levels of female labour force participation and good childcare provision also tend to have higher fertility rates. The effectiveness of family policies can be assessed by looking at poverty risks for different family types or by looking at fertility rates which could be seen as an indicator of whether people feel able to realise their desire to have children. Fertility rates appear not to be correlated to the generosity of cash benefits, but they are positively correlated to the availability of childcare for the youngest children and to female employment rate: countries with a high level of childcare provision and high levels of female labour force participation also tend to have higher fertility rates. A reorientation of family support towards measures that facilitate the reconciliation of work and family life thus seems to be desirable both from the point of view of reducing poverty risks and raising fertility rates. In addition, it boosts employment and enhances the EU's ability to cope with the needs of an ageing population. The EU promotes better policies in the Member States by implementing the roadmap for gender equality, through the open method of coordination in the area of social protection and social inclusion and through the establishment of a European Alliance for Families.

The baby-boom started 60 years ago; today, these cohorts start retiring in large numbers. Chapter 3 looks at opportunities and needs in an ageing society, focussing in particular on the ageing baby-boomers and their potential for contributing to the economy and society. The importance of such an analysis was highlighted in the Commission's Renewed Social Agenda of July 2008. The EU population pyramid clearly shows an increase in cohort sizes just after the end of World War II, marking the start of the baby boom. This was 60 years ago and the first of these large cohorts born over a period of 20-30 years are now beginning to retire. This marks a turning point in the demographic development of the European Union, and ageing is no longer something that will happen at some point in the distant future. Over the past decade, both the population of working age (20-59 years) and the population aged 60 years and above had been growing by 1 to 1.5 million people per year on average. From now on, the population aged 60 years and above will be growing by 2 million people every year for the next 25 years. The growth of the working-age population is slowing down fast and will stop altogether in about 6 years; from then on, this segment of the population will be shrinking by 1 to 1.5 million people each year.

Employment rates at 60 are ten percentage points higher than in 2000, but there is still much room for improvement. In 2007, around 50% of men and 40% of women were still in employment at the age of 60. These rates are low, indicating that the ageing baby-boomers constitute a major potential for increasing Europe's labour force. Nevertheless, this represents an increase by 10 percentage points compared to the year 2000 and shows that the trend towards earlier retirement has been reversed. It is also one of the clearest indications that the Lisbon strategy, which aims among other things at promoting employment of older persons, is working. Employment after the age of 65, the typical statutory retirement age in many Member States, is very rare: only about 13% of men aged 65-69 years and 7% of women are still in employment. Part-time working could be a good way of achieving a gradual transition from work to retirement, but only about 11% of men aged 55-64 work part time and 38% of women. Thus, for men, the typical transition still seems to be from full-time employment to full-time retirement, whereas for many women in this age class part-time working may have been a way back into the labour market when their children required less time. By contrast, after the age of 65, part-time work is a very common form of employment for the few women and men who are still on the labour market: 47% of employed men and 61% of women aged 65+ worked part-time in 2007. Future cohorts of older workers will be better educated and have better ICT skills, thus raising their chances of staying in employment... Education and skills, and in particular digital literacy, remain an obstacle to increased labour force participation of people in their fifties and sixties. However, the situation can be expected to improve as future cohorts entering this age class are characterised by a higher level of educational attainment and much greater familiarity with computers and the internet. In 2007, 57% of people aged 55-64 had never used the

internet; in the age group 45-54, the proportion was 39% and for people aged 35-44 it was 28%. A much smaller proportion of people have also reached a lower level of educational attainment in the younger cohorts than in the older ones. The increase in the level of educational attainment is particularly strong for women: 34% of women aged 25-29 have tertiary education, more than twice the proportion for women aged 55-59 (16%); for men, the progression is much less pronounced: from 21% in the age group 55-59 to 25% in the age group 25-29.

...but caring obligations and poor incentives in taxbenefit systems could remain obstacles to the employment of older workers. While higher levels of educational attainment can be expected to allow more workers to stay longer on the labour market, further analysis is required on the health status of older workers and to ascertain whether enough is being done to update skills of ageing workers in accordance with current needs in the labour market. Moreover, caring obligations towards grandchildren or dependent adults could represent an obstacle to increased employment, particularly of women in their fifties and sixties: at this age, they may be expected to care for grandchildren and their ageing parents. Finally, tax and benefit systems will have to provide incentives for staying longer on the labour market. All these issues will be given further attention, notably within the framework of the Lisbon Strategy and the Open Method of Coordination on social protection and social inclusion. Even after retirement, older people can make a major contribution to society... The ageing baby-boomers not only have the potential of making a major contribution to the economy through their participation in the labour force, but older people also engage in a wide range of social activities, organised by religious, political, trade union, charitable or recreational organisations, or informally by helping relatives and other people in the community. According to a special module on participation in the European Survey on Income and Living Conditions (EU-SILC), people over the age of 65 tend to be more active in church and religious activities than people under the age of 65, and less active in political, trade union and recreational organisations. However, the proportion of older people participating in such activities is not very high: around one quarter in church and religious activities, 3% in political parties and trade unions and around 20% in recreational groups and organisations. ...by looking after grandchildren or other relatives in need or as volunteers in their communities. The Survey on Health and Retirement in Europe (SHARE) covers fewer countries, but provides a more detailed picture of social participation of people over the age of 50. About two in five grandparents in the countries covered by SHARE provide care to their grandchildren, one in five grandfathers and almost one in four grandmothers almost weekly or more often. There are huge differences across countries as far as activities such as volunteering, informal helping, caring for other adults and other social activities in clubs, political or community organisations are concerned. Older people are most likely to be involved in these activities in the Netherlands, Sweden and Denmark, whereas participation levels are lowest in the Southern and Central European countries taking part in the Survey.

There are major differences in the social activities of older workers across countries more so than across socioeconomic groups in a given country. These cultural differences across Member States appear more important than personal characteristics which, however, are also strong determinants of social involvement of the over fifties. Men tend to be slightly more involved than women in most activities apart from caring; the highly educated are much more involved than those with a low level of education; and participation generally declines with age, although, in some countries with a high level of volunteering, people aged 65-74 are more active than people aged 50-64. The fact that there such large crosscountry differences suggests that, in countries with low levels of participation, any policy attempts to promote a greater active involvement of older people in society may need to foster first of all a strong culture of social participation.

Rapid ageing requires adequate policy responses: opportunities to stay active on the labour market and in society; access to goods and services that preserve older people's autonomy; solidarity with the dependent and protection of their dignity. The ageing of the baby-boomers has a number of policy implications. The need to promote the employment of older workers has already been recognised within the Lisbon Strategy. In various Member States, policy makers are also trying to promote voluntary work by older people. Apart from creating opportunities for the active participation of older people, policy makers will have to develop policies aimed at preserving the autonomy of older people. This comprises financial autonomy as well as physical autonomy thanks to adapted housing, transport and access to services that allow older people to stay in their own homes for as long as possible. For those older people who have become highly dependent on the help of others, a coordinated offer of health and longterm care services must be developed. Member States can prepare for demographic change by acting in five key areas. The fourth chapter presents information on the preparedness of the EU and its Member States for demographic change. This is done using some key indicators relating to each of the five policy areas in which Member States can take action to tackle the challenges of demographic change. These indicators are also presented in country sheets which show at a glance where a given Member State stands with regard to demographic challenges and policy responses, comparing it to the EU average and the best performing countries.

They can create better conditions for families and mothers in particular, thus contributing to higher fertility rates while at the same time offering better opportunities to women. With a view to identifying the conditions needed for Europe's demographic renewal, the chapter illustrates the diversity of Member States with regard to their spending on family benefits (also discussed in chapter 3). It also shows that, in 2006, a majority of Member States still fell short of the targets set by the European Council in Barcelona, namely to ensure access to formal childcare for one third of children under the age of three and for 90% of children aged 3-6. Among the countries with the lowest levels of childcare provision are most Central and Eastern European countries. Almost 90% of men with children under the age of six are in employment, compared to less than 60% of women. Moreover, a significant proportion nearly one third of all

women, are working part-time. Women's disproportionate career sacrifices (relative to men's) for the needs of their families also show up in the large gender pay gap: women earn 15% less per hour worked than men.

Member States can raise labour force participation, thus creating a better balance between the active and the retired. A society's ability to cope with an ageing population does not depend directly on the old-age dependency ratio, i.e. the number of people over 65 in relation to people aged 15-64 ('working-age population'). The key question is how many inactive people and people with expensive health and long-term care needs have to be supported by the active population, which is much smaller than the total population aged 15-64. Just under two-thirds (65.4% in 2007) of the working-age population are actually in employment, several percentage points below the target set within the Lisbon Strategy for the year 2010. Progress towards this target has been slow: just over three percentage points since the target was set in 2000, meaning that, so far, less than half of the gap between the starting position and the target has been bridged.

In about ten years, the potential for further employment growth will be exhausted; productivity will become the main engine of growth... While there is still good potential for increasing employment through increased labour force participation, notably of women and older workers, it can be expected that within about one decade, the decline of the working-age population will be such notably due to the retirement of the baby-boomers that rising employment rates will no longer be sufficient to compensate for this decline. From then on, the source of economic growth will have to be increases in productivity, which need to be achieved through investment in human and physical capital and in innovation. The chapter provides some indications on how well Europe's human potential is used. Labour productivity per hour worked is a key indicator in this regard. Between 2003 and 2007, it grew by between 1 and 1.7% annually. The fastest progress took place in the new Member States, which are catching up with the EU average, albeit from a very low level. ...but this requires more investment in human capital... Productivity growth depends notably on the level of educational attainment. 13% of women aged 18-24 and 17% of men have only attained lower secondary education or less and are defined as early school leavers. This represents a modest improvement compared to the beginning of this decade, but remains far off the target set for 2010, namely to bring this proportion down to 10% at most. Southern European countries are facing particularly strong challenges in this regard, whereas the Central and Eastern European countries tend to be among the best performers, which is also reflected in a high proportion of young people completing at least upper secondary education. However, these results are not clearly correlated to public spending on education. ...and innovation. In both areas, there is considerable scope for improvement. Finally, future productivity growth also depends on technological progress which depends on universities educating graduates capable of engaging in research and on devoting money to research and development. The number of university graduates relative to the population in their age class differs widely across Member States. It is

about two-and-a-half times higher in the UK, Denmark and the Netherlands than in Germany and Austria. The new Member States are spread out across the country ranking. With regard to the proportion of GDP devoted to research and development, the new Member States are trailing far behind. Most of them spend less than 1% on R&D, compared to the EU-27 level of 1.84% in 2006. This is far below the target set for 2010 of 3% of GDP. Moreover, the level of R&D spending has not risen since the beginning of the decade.

Immigration can alleviate labour market shortages, but Member States differ in their capacity to integrate immigrants in their labour markets. Immigration can help alleviate labour market shortages due to a declining working-age population. The EU has been receiving an unprecedented number of migrants over recent years. Over the past 13 years (1995-2007), the population of EU-27 increased by nearly 15.5 million people due to net migration, 4.5 million during the first seven years and 11 million during the last six years of this period. The countries that attracted the largest numbers of migrants were Spain, Italy, Germany and the UK. Several new Member States experienced net emigration during that period, but this flow appears to have come to a halt in more recent years. Thus, immigration plays a very different role in national population dynamics across the Member States. Considerable differences also exist with regard to the integration of immigrants into the labour market. The employment rate of male nonEU-27 nationals is 2.5 percentage points lower than that of nationals; for women, the gap is more than ten percentage points. However, there are considerable differences across Member States, and in a number of them the gap is actually in favour of non-nationals. Member States can also consolidate their public finances and thus reduce the need to allocate public spending to interest payments. The fifth area in which Member States can respond to the challenges of demographic change are public finances, which was addressed in the Communication on The long-term sustainability of public finances in the EU 5

. Moreover, ensuring progress towards sustainable public

finances is was a key element in the 2005 reform of the EU fiscal framework and the Stability and Growth Pact. This Communication provided an assessment of the scale and scope of the fiscal sustainability challenge each Member State is facing in view of an ageing population. The increase in the number of older people will create additional public expenditure demands for pensions, health and long-term care

. Reforms

of social protection systems, making them more efficient and encouraging older workers to stay longer on the labour market, can curb the increase in expenditure to some extent. Governments can prepare for these needs of an ageing society by reducing their public debt and hence the amount of tax revenue they need to allocate for interest payments. In 2007, government debt amounted to 60% of annual GDP in EU-27, the lowest level for the past twelve years. The total amount of budget

5 COM(2006) 574 adopted on 12 October 2006. A new assessment of public finance sustainability is to be released in the autumn of 2009. 6

See Economic Policy Committee and European Commission (2006), The impact of ageing on public expenditure: projections for the EU25 Member States on pensions, health care, long-term care, education and unemployment transfers (2004-2050) , European Economy, Special report, No.1/2006

deficits of the Member States represented 1% of GDP for EU-27, down from 3% 5 years earlier. The situation varies, however, considerably across Member States and ranges from a budget surplus of over 5% of GDP to as large a deficit. The amount of government debt ranges from more than 100% of GDP to less than 10%, and this is also reflected in interest payments: in 2007, Italy had to use more than 10% of its public spending on debt interest.

I NTRODUCTION 7

The Commission's Renewed Social Agenda identified population ageing, alongside with

technological progress and globalisation, as one of the key drivers of societal change in Europe. The prospect of ageing populations has been discussed for a long time. Today, as the first baby-boomers turn 60, it is no longer something that will happen at some point in the distant future. 60 years ago, the number of babies born rose sharply and remained high for about 20 to 30 years. Now the first of these large cohorts born over a period of 20 to 30 years are beginning to retire. This marks a turning point in the demographic development of the European Union and makes it all the more important to consider the policy responses that are required by this major change.

This report has been placed under the theme of Meeting Social eeds in an Ageing Society. It thus follows up on the Commission's commitment in the Renewed Social Agenda to look at the needs of an ageing population. However, the focus is not just on needs, which would inevitably strengthen the perception of ageing as a burden. There are numerous opportunities for tackling the challenges of ageing and for modernising European societies, creating better living conditions for people of all ages.

Indeed, as the Commission stressed in its communication presented in October 2006 on The Demographic Future of Europe -- From Challenge to Opportunity

, Europe can envisage her

demographic future with confidence. Such confidence is founded on the recognition that population ageing is above all the result of economic, social and medical progress, as well as greater control over the timing and number of children people wish to have. It was also based on the realisation that Europe faces significant opportunities to respond to the challenges of demographic change in five key areas:

Better support for families; Promoting employment;

Reforms to raise productivity and economic performance; Immigration and integration of migrants; Sustainable public finances.

However, major reforms and decisive action are necessary, and the Communication underlined that there is only a short window of opportunity of about ten years during which further employment growth would remain possible. Increasing the number of highly productive and high-quality jobs is the key to ensuring that Europe's economy and societies will be able to meet the needs of ageing populations.

It falls above all to each Member State to develop the right policy mix in response to demographic change, and each Member State faces somewhat different challenges and has different sets of opportunities. The Communication of 2006 offered a broad reference

See Renewed Social Agenda: Opportunities, Access and Solidarity in 21st Century Europe , COM(2008) 412 of 2 July 2008. 8 COM(2006) 571, adopted on 12 October 2006.

framework to help Member States develop their specific policy mixes, the success of which will be in the interest of the European Union as a whole. It also emphasised that the Lisbon Strategy for Growth and Jobs, alongside the open method of coordination in the area of social protection and inclusion, offers the appropriate framework for conducting the reforms that are necessary to tackle demographic change.

The Communication announced that every two years, the Commission would hold a European Forum on Demography to take stock of the latest demographic developments and to review where the European Union and the Member States stand in responding to demographic change. The first Forum took place on 30-31 October 2006, the second on 24-25 November 2008.

A tool for assessing where we stand in relation to the demographic challenge

The purpose of this second report on demography in Europe is to provide the latest facts and figures that are needed for an informed debate with the stakeholders taking part in the Forum and in particular with the group of government experts on demography, which was involved in the conception of the present report.

As far as possible, data are provided for each Member State, allowing policy makers and stakeholders to compare their own country's situation with that of others, to understand the specificity of their country and, possibly, to identify countries that provide interesting experiences from which lessons could be learned. In so doing, the report responds to the request from Member States who want to learn from the variety of national experiences across the European Union. The data provided in this report concern the latest demographic trends and projections (chapter 1). In particular, the report discusses the latest Eurostat population projections which present a more optimistic outlook, foreseeing a slight growth of the population as opposed to a slight decline according to the previous projections. However, the challenge of population ageing remains huge and hardly changed compared to the earlier projections.

In addition to demographic data, the report also examines some key indicators showing where the EU as a whole and individual Member States stand in their preparation for demographic change (chapter 4). This analysis, which is one of the commitments entered into by the Commission in the communication of October 2006, follows the five key areas for action. It helps policy makers in the Member States identify those areas where they face the most important opportunities for tackling the demographic challenge.

Chapter 4 can only provide a very cursory overview of the challenges and opportunities in these five areas, each of which is the subject of more in-depth scrutiny in the context of the Lisbon Strategy. The aim of this report is to present a broad picture of where individual Member States and the EU as a whole stand in the five key areas for tackling the challenges of demographic change. In addition, the report aims at complementing information that is already available thanks to ongoing policy coordination processes, by looking at two specific areas, namely the needs of families and older people in a changing demographic context.

Two priority areas for policy review: the needs of families...

The 2008 demography report focuses in particular on two issues that have received much attention following the adoption of the communication on Europe's demographic future: the

modernisation of family policies and opportunities for enhancing the contribution of older 10

people to the economy and society .

As far as family policies are concerned, the report examines in chapter 2 how the reality of family life has changed over recent decades. Couples have become less stable and choose to have children at a higher age, often without being married. Women today face much better opportunities on the labour market and, thanks to the rapid progression of their educational attainment levels, are much better equipped to seize those opportunities. In this context, family policies oriented towards the traditional male breadwinner model are becoming less effective in securing good living conditions for families and children. The chapter discusses the policy implications of the transformation of family life that has taken place in Europe and emphasises in particular the importance of reconciliation policies.

...and older people

Chapter 3 is devoted to opportunities and needs in a society that is ageing at an accelerating pace. Over the past decade, both the population of working age (20-59 years) and the population aged 60 years and above had been growing by 1 to 1.5 million people per year on average. From now on, the population aged 60 years and above will be growing by 2 million people every year for the next 25 years. The growth of the working-age population is slowing down fast and will stop altogether in about six years. From then on, this segment of the population will be shrinking by 1 to 1.5 million people each year.

Thus the baby-boom cohorts will no longer boost the working-age population, as they did over the past decades, but will start increasing the population of pensioners. In about fifteen to twenty years, these same cohorts will lead to a strong expansion of needs for health and social care services. The ageing of the baby boom will put solidarity between the generations to a difficult test.

The potential contribution of the ageing baby boomers

This report looks at the potentials for increased participation of older people, on the labour market and as active members of their communities, notably through volunteering. Mobilising this potential is all the more urgent as much larger cohorts are now entering their 60s. At present, only about 40 % of men and 30 % of women were still in employment at the age of 60. Yet, most people in this age group are still fit and capable of contributing to the economy and society. The Lisbon Strategy is already producing tangible results: Employment rates of people aged 55-64 are rising, reversing the past trend towards ever earlier retirement, but more needs to be done. Opinion surveys also indicate a great willingness to participate in community work or volunteering after retirement. This, too, represents a major opportunity for social progress, but figures on actual engagement fall far short of this declared willingness to volunteer. Clearly, more and better opportunities for employment and voluntary engagement of older people are needed.

Another challenge for policy makers to ensure that older people have access to the goods and services they need. This requires adequate incomes as well as a supply of goods and services that are adapted to the specific needs to older people, allowing them to remain autonomous

See the Communication from the Commission Promoting solidarity between the generations, COM(2007) 244. 10 Council resolution of February 2007, DOC 6216/1/07.

and live in their own household for as long as possible. While much work on incomes in old age has already been done in the context of the Open Method of Coordination on social protection and social inclusion, there is a lack of data on the barriers that older people may be facing and the goods and services that could help overcome those barriers. However, new surveys will soon begin to fill this knowledge gap, and in future demography reports it will be possible to deepen the analysis.

An emerging challenge: protecting the dignity of frail older people

A fundamental issue that has recently been put on the European agenda is the protection of the dignity of frail older people who are often victims of neglect and abuse

. A major

European conference took place on 17 March 2008 in Brussels and highlighted the seriousness of the issue, albeit based on very limited information from a few Member States only. The chapter on needs and opportunities in an ageing society discusses some of the policy challenges that need to be tackled and indicates some ways in which the European Union can support policy makers in the Member States.

Help us improve future reports...

Comments and suggestions on this report will be gratefully received and should be sent to:

Unit E1 (Social and Demographic Analysis)

Directorate-General for Employment, Social Affairs and Equal Opportunities European Commission

B-1049 Brussels EMPL-E1-COURRIER@ec.europa.eu

11 See notably the exploratory opinion of the European Economic and Social Committee on Elder Abuse , Brussels, 24 October 2007 (CESE 1465/2007).

1.

E UROPE ' S C URRENT D EMOGRAPHY AND F UTURE T RENDS

The most common way of representing a country's population is by using what is known as a 'population pyramid', which displays the population by age and sex. However, the shape of the pyramid that this representation of population structure typically took in the past has long been replaced by very different contours, much narrower at the base as a result of a declining number of births over recent decades. The pyramid can result from birth rates high above the replacement level or from high mortality at any age. While low birth rates have become a concern in many Member States, a return to pyramid-shaped population structure would not be desirable, because high birth rates implies explosive population growth, whereas high mortality at any age implies that most people will not be able to enjoy a long life. The ideal shape for the population structure would therefore be a pillar, which narrows only at the very top as people die of old age.

Figure 1.1 on the following pages presents current and projected population structures in 2060 for the EU as a whole and for each Member State. The population pyramids show a great diversity across the EU, both for the present and for the future, reflecting the political and social history of each country. The impact of low birth rates and high male mortality during war years is visible in a number of countries.

In the EU-27 pyramid, the baby boom, which peaked in the mid-1960s, is clearly visible. Most countries experienced a period with high birth rates, typically some 40 to 50 years ago in the EU-15 countries

, and a decade or two later in the former communist Member States.

The effects of the peaks in birth rates also tend to be felt some 20 to 30 years later when the children born during a baby boom have their own children, as can be seen very clearly in the case of the Czech Republic. Most Member States are, however, experiencing a declining number of births and thus a narrowing of the base of their population pyramid.

Projections for population structure in 2060 suggest that the EU-12 as well as the Southern

Member States and Germany might have almost inverted population pyramids with ever small birth cohorts. Greater life expectancy will result in much larger cohorts aged 60 and above, particularly for women. A few Member States, the Nordic countries, Ireland, the UK and France, are expected to move towards the pillar shape resulting from birth rates close to replacement level and low mortality (or high life expectancy).

These current and projected future population pyramids are shaped by births, deaths and a third key factor, namely migration. These factors are considered in the present chapter, which examines how they shape the composition and structure of Europe's population.

EU-15 will be used to refer to the Member States of the EU before the 2004 enlargement.

13 EU-12 is used to indicate the Member States which acceded in 2004 and 2007. EU-10 refers only to the Member States that acceded in 2004.

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) BE

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) CZ

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) DE

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) IE

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) ES

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) IT

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) LV

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) LU

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) MT

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) AT

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) PT

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) SI

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) FI

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060) UK

Men Women 100

g 50 A

1,00 0,75 0,50 0,25 0,00 0,25 0,50 0,75 1,00 Per cent

Men (2008) Women (2008) Men (2060) Women (2060)

Source: Eurostat, EUROPOP2008 convergence scenario

1.1. Trends in birth rates

In 2005, around 5.1 million children were born in the 27 Member States of the EU, compared to nearly 7.6 million in 1965. The absolute size of birth cohorts may fluctuate significantly, as can be seen from the population pyramids presented above, requiring adaptations to childcare, and education infrastructures, in particular, to accommodate variations in the number of births.

The number of births may vary due to changes in the size of cohorts of women of childbearing age: a baby boom leads to a secondary baby boom 20 to 40 years later. However, a major factor in the decline of the number of births has been a profound change in behaviour: European women have fewer babies than in the 1960s, as demonstrated by the indicator for completed fertility, which refers to the mean number of children born to women of a given generation at the end of their childbearing years (assumed to range from 15 to 49 years).

Table 1.1 shows that women born in 1935 had on average 0.4 children more than women born 20 years later. The figures for women born in 1965 are not fully comparable since some of these women had at that time not yet reached the end of their reproductive life span, but those for 1955 can be compared. The table also reveals other changes in reproductive behaviour. In particular, the mean age of women at the birth of their first child has risen, typically by two to three years in the countries for which data are available. For this indicator, the figures in table1.1 for the cohort born in 1965 can be considered as minimum values, which are expected to increase as a result of births that have not yet been recorded.

Voluntary childlessness (assuming that it does not result from an untreatable medical condition) is another aspect of fertility behaviour affecting population structure. For most of the countries for which data are available, table 1.1 shows how the proportion of childless women has increased slightly. The figures for women born in 1965 may underestimate their completed fertility rate, because they could still have a child.

The completed fertility indicator has limitations since it reveals changes in fertility behaviour only at the end of the reproductive life span of a cohort, and hence with a considerable time lag. Demographers have developed a way of estimating current fertility using an indicator referred to as the total fertility rate. The TFR is the mean number of children that would be born alive to a woman during her lifetime if she were to pass through her childbearing years conforming to the fertility rates by age of a given year. It is an estimate based on childbearing probabilities currently observed for women of different cohorts. It therefore represents the completed fertility of a hypothetical generation, computed by adding the fertility rates by age for women in a given year (the number of women at each age is assumed to be the same). The total fertility rate is also used to indicate replacement level fertility; in more developed countries, a rate of 2.1 is considered to be necessary to replace fully an existing generation. Table 1.2 presents trends in the TFR in EU Member States.

Table 1.1: Fertility indicators, selected birth cohorts of women, 1935-65

Total cohort fertility rate Mean age mother at first % women without children Number of birth children per woman, cohort

1960(%)

1935 1945 1955 1965* 1935 1945 1955 1965* 1935 1945 1955 1965* 0 1 2 3+

EU-25** 2.37 2.11 1.94 1.77 26.3 15.7 15 18 40 27 EU-15** 2.36 2.10 1.90 1.73 26.9 16.1 16 18 39 26 EU-10** 2.40 2.16 2.10 1.96 23.2 13.4 10 17 44 29 AT 2.45 1.96 1.77 1.64 25.3 21.0 17 23 39 15

BE 2.27 1.93 1.83 9.4 8.7 10.5 17 28 34 20

CY 2.31 2.56 CZ 2.12 2.03 2.07 1.93 22.5 7.0 6 15 55 23 DK 2.35 2.06 1.84 1.94 23.3 25.0 27.1 8.1 13.1 13.0 10 22 43 25

EE 1.85 2.00 1.87 FI 2.29 1.88 1.90 1.91 24.4 25.2 27.0 14.2 18.2 20.0 19 14 36 31

FR 2.57 2.22 2.13 2.02 24.5 24.0 24.7 26.3 10.3 7.0 7.8 28 18 32 22 DE 2.16 1.80 1.67 1.53 30 19 35 15 GR 1.98 2.01 1.75 25.2 18.0 11 16 52 22

HU 1.99 1.90 1.94 1.97 23.0 10.0 7 20 49 23

IE 3.52 3.27 2.67 2.18 25.7 24.9 25.1 27.4 4.5 6.2 13.2 21.0 16 10 28 46

IT 2.28 2.07 1.80 1.49 25.3 24.4 24.5 27.0 13.1 10.2 11.1 20.0 15 25 43 17

LV 1.84 1.77

LT 1.97 1.94 1.72 LU 1.82 1.69 1.82 MT 2.00 NL 2.49 2.00 1.87 1.77 25.5 24.5 26.3 28.4 12.5 12.4 17.4 19.0 18 15 42 25 PL 2.60 2.27 2.17 2.00 23.3 16.0 11 17 39 33

PT 2.88 2.42 2.04 1.82 25.2 24.6 23.9 25.2 4.1 4.9 7.5 5.0 6 30 45 21 SI 2.07 1.83 1.96 1.77 23.7 9.0 4 26 54 16 SK 2.72 2.38 2.22 2.04 22.7 11.0 10 13 45 32

ES 2.14 1.98 2.03 1.98 24.0 25.4 26.8 6.1 12.8 13.0 13 15 41 31

SE 2.01 1.89 9.8 17.2 UK 2.03 1.83 3 BG 2.00 1.78 1.92 1.88 23.9 5.0 5 22 52 21

HR 2.38 2.44 2.28 1.91 22.5 12.0 8 24 39 28

RO 4.48 3.97 TR 4.48 3.97 * Some women in this birth cohort may still have been in their reproductive period ** Estimates for the 10 Member States that joined in 2004

Source: Demography Monitor 2005, NIDI Netherlands, European Observatory on Demography and the Social Situation- Demography Network, European Commission.

Table 1.2: Total Fertility Rates in the EU

1960/64* 1970/74* 1980/84* 1990/94* 1995/99* 2000/04* 2006 EU-27** 2.64 2.23 1.79 1.56 1.47 1.46 1.53

BE*** 2.64 2.07 1.61 1.62 1.58 : 1.74 BG 2.23 2.16 2.01 1.57 1.18 1.24 1.37 CZ 2.22 2.14 2.01 1.72 1.18 1.17 1.33 DK 2.58 1.97 1.44 1.73 1.76 1.76 1.83 DE 2.46 1.77 1.48 1.32 1.33 1.35 1.32 EE : 2.13 2.12 1.67 1.33 1.39 1.55

IE 3.91 3.84 2.92 1.99 1.89 1.93 1.90 GR 2.25 2.33 2.02 1.37 1.27 1.27 1.39 ES 2.86 2.87 1.94 1.30 1.17 1.27 1.38 FR 2.83 2.36 1.88 1.72 1.80 1.90 2.00

IT*** 2.50 2.37 1.55 1.28 1.21 1.27 1.35 CY 3.47 2.38 2.46 2.35 1.85 1.54 1.47 LV 2.01 2.01 1.70 1.18 1.24 1.35

LT 2.57 2.28 2.04 1.86 1.49 1.29 1.31 LU 2.33 1.77 1.48 1.65 1.72 1.66 1.65 HU 1.88 2.01 1.82 1.77 1.40 1.30 1.34 MT 3.16 2.21 1.98 2.02 : : 1.41

NL 3.17 2.15 1.52 1.59 1.58 1.73 1.70 AT 2.78 2.08 1.61 1.49 1.39 1.38 1.40 PL 2.76 2.24 2.33 1.93 1.51 1.27 1.27 PT 3.16 2.71 2.05 1.53 1.46 1.46 1.35 RO 2.10 2.65 2.18 1.55 1.39 1.30 1.31 SI 2.25 2.14 1.91 1.38 1.25 1.23 1.31 SK 2.93 2.50 2.29 1.94 1.42 1.22 1.24 FI 2.68 1.64 1.68 1.82 1.75 1.75 1.84 SE 2.30 1.90 1.64 2.04 1.57 1.64 1.85 UK 2.86 2.20 1.81 1.78 1.71 1.68 1.84 *average for 5 year period calculated by NIDI **EU-27 figures for 2000/04, 2006 are estimates due to missing values for Belgium, Italy and Malta *** Missing Belgian and Italian figures inserted from the 2008 European demographic data sheet, see

www.populationeurope.org

Source: NIDI, UN data for the years before 1990 and Eurostat demographic data for the years thereafter.

Table 1.2 shows that the total fertility rate has declined steeply since the 1960s and 1970s to a level that is far below replacement level: around 1.5 for the EU-27, with values close to, or even below, 1.3 in a number of Member States, particularly in the Central and Eastern Member States and Germany. Fertility is also low in the Southern Member States. The countries with the highest fertility rates are France, the UK, Ireland, and the Nordic countries.

Since the 1980s, there has also been a significant change in the timing of births. In 2003, women in EU-15 Member States tended to have their first child around three years later in life than in the early 1960s and 1980s (see Table 1.3). By contrast, a similarly strong increase has not occurred in EU-10 Member States, where women tended to be 2.7 years younger at the birth of their first child than first-time mothers in EU-15.

Table 1.3: Postponement of childbirth in EU-15 and EU-10

Mean age of mothers at first birth

1960 1980 2003 EU-25 24.4 24.6 27.5 EU-15 24.5 25.0 28.0 EU-10 24.0 23.1 25.3

Source: Demography Monitor 2005, NIDI Netherlands, European Observatory on Demography and the Social Situation- Demography Network, European Commission.

More recent data (see Table 1.4) suggest that postponement of childbirth is now also happening in the EU-10 countries which appear to be following the pattern in EU-15. In many Member States, the average age of mothers at the birth of their first child appears to have increased by around five years since 1980.

Table 1.4: Mean age of women at first child birth

1980 1995 2000 2005 2006* BE : 27.3 : : 28.7

BG 21.9 22.4 23.5 24.7 26.2 CZ 22.4 23.3 24.9 26.6 28.9 DK 24.6 27.4 27.3 28.4 30.3 DE : 27.5 28.2 29.1 29.6

EE 23.2 23.0 24.0 25.2 28.4 IE 25.5 27.3 27.7 : 31.0

GR 24.1 26.6 28.0 28.5 29.8 ES 25.0 28.4 29.1 29.3 30.9 FR 25.0 28.1 27.9 28.6 29.8 IT 25.0 28.0 : : 30.9 CY 23.8 : : 27.5 29.8

LV 22.9 : 24.0 25. 27.8

LT 23.8 23.1 23.9 24.9 27.7 LU : 27.4 28.3 29.0 30.0

HU 22.5 23.8 25.1 26.7 28.6 MT : : : : 28.4

NL 25.7 28.4 28.6 28.9 30.6 AT : 25.6 26.4 27.2 29.2

PL 23.4 23.8 24.5 25.8 28.3 PT 24.0 25.8 26.5 27.4 29.4 RO 22.4 22.9 23.7 24.8 26.8 SI 22.8 24.9 26.5 27.7 29.6 SK 22.7 : 24.2 25.7 27.9

FI 25.6 27.2 27.4 27.9 30.0 SE 25.3 27.2 27.9 28.7 30.5 UK 24.7 : 29.1 29.8 29.2

*Figure for 2006 is the estimated mean age at childbearing, referring to all children not just the first child. Source: 1980 NIDI, 1995, 2000, 2005 Eurostat, 2006 estimated starting values of Eurostat, EUROPOP2008 convergence scenario.

A higher age at the birth of the first child reduces the time left before the end of the reproductive life span. This could in itself reduce fertility rates if an increasing number of women find themselves confronted with problems of biological infertility when they want to have further children. Figure 1.2 shows, however, that the countries with the highest fertility rates also display high average ages of mothers at the birth of their first child.

Figure 1.2: Total fertility and the estimated mean age at first birth, 2006

Total Fertility Rate and Mean Age at First Birth, 2006

2,0 FR 1,9 IE SE

UK FI DK

1,8

1,7 NL LU

F 1,6 T

1,5

1,4 AT GR BG ES

LV PT IT HU

CZ DE

RO LT SI 1,3

1,2

26,0 26,5 27,0 27,5 28,0 28,5 29,0 29,5 30,0 30,5 31,0 Mean age

Source: See Tables 1.2, 1.4.

As more and more couples are postponing childbirth, biological fertility problems are likely to become an obstacle in a growing number of cases. The development of assisted reproductive technology (ART) such as in vitro fertilisation may make it possible for couples to have children at a later age. However, offering wider access to ART is unlikely to have a significant incidence on demographic trends (see Box 1.1).

Box 1.1: Can assisted reproductive technology (ART) promote higher fertility?

It has been suggested that making ART widely available to all those who fail

to conceive within a timeframe of one year, could raise the present fertility rate with 0.20 children per woman in the UK and 0.17 in Denmark. These are considerable effects that would help to offset population ageing.

However to reach this goal the capacity for in vitro fertilisation (IVF) would have to rise 5 to 10-fold compared the present capacity in these countries. Indeed, if such treatment is offered after one year of unsuccessful attempts to conceive (and was widely accepted), many women who would have spontaneously conceived after the first year would receive the treatment. An early application of ART treatments to a large group, with no clearly identified cause of infertility, therefore mainly results in pregnancies occurring slightly earlier than without the treatment. But it would expose them also to the

14 Hoorens S. et al. (2007) Can assisted reproductive technologies help to offset ageing? An assessment of the demographic impact of ART in Denmark and the UK , Human Reproduction 2007, 22: 2471-2475 EN

complications and the side effects of the treatment . Although there would be a

positive effect on fertility levels, this would be mainly due to the many babies from multiple pregnancies that often result from ART and lead to many more complications than from singletons after IVF. If all IVF pregnancies would lead to just one baby the net effect on fertility rates when treating patients already after one year would be negligible. The impact of IVF as a policy measure to promote fertility rates in order to offset population ageing is therefore highly unlikely.

Between 1997 and 2002, the number of ART treatment cycles has increased by 59%, from approximately 204 000 to 324 000, respectively, with (IVF) and intracytoplasmic sperm injection being the most prevalent treatments. As a result of the total cycles of ART in 2002, approximately 49 000 births occurred across 25 European countries, i.e. around 1% of all births. However, in Denmark in 2002, 4.2% of children were born following ART treatment

Postponement of childbirths also makes it more difficult to estimate total fertility rates. They are based on the assumption that the probability for a woman to have a child when she reaches a certain age will be the same as the probability of giving birth for women who are in this age group today. Postponement from, say, 25 to 30 years of age means that the probability of giving birth at the ages 25 to 29 will fall. This will lower the TFR indicator, even though the women who are postponing may have exactly the same number of children at the end of their reproductive life span. Postponement thus depresses the TFR until the process has come to an end.

It is likely that some of the lowest TFR values in the EU are in fact the result of postponement. The Vienna Institute of Demography has tried to adjust TFR for postponement effects

. Figures presented in Table 1.5 suggest that actual fertility could

be almost 0.2 children per woman higher than the unadjusted TFR. The adjustment is highest in the EU-12 countries, typically between 0.3 and above 0.4, and smallest in countries such as Belgium, Greece, Spain, France, Italy, the Netherlands, Finland and Sweden, where the postponement process seems to be nearing the end.

15 Te Velde E. et al. (2008) Can assisted reproductive technologies help to offset population ageing? Human Reproduction 2008, advance access published June 21.

16 Sorenson C. and P. Mladovsky (2006), Assisted reproduction technologies in Europe: an overview , Research Note LSE, European Observatory on Demography and the Social Situation- Health Network, European Commission 17

See the 2008 VID demography data sheet, http://www.oeaw.ac.at/vid/popeurope EN

Table 1.5: Period, Tempo Adjusted and Cohort Fertility Rates

Period TFR Tempo Adjusted Gap 1965 Cohort fertility 2006 period TFR mean of (2)-(1) rate 2003-2005*)

(1) (2) (3) (4)

EU-27 1.53 1.72 0.19 1.79 BE 1.74 1.86 0.12 1.79 BG 1.38 1.70 0.32 1.53 CZ 1.33 1.76 0.43 1.93 DK 1.85 2.00 0.15 1.89 DE 1.33 1.59 0.26 1.55 EE 1.55 1.85 0.30 1.95 IE 1.90 2.17 0.27 2.32 GR 1.40 1.52 0.12 1.77 ES 1.38 1.39 0.01 1.61 FR 1.98 2.07 0.09 2.03 IT 1.35 1.48 0.13 1.50 CY 1.44 1.79 0.35 2.57 LV 1.35 1.59 0.24 1.84 LT 1.31 1.68 0.37 1.74 LU 1.64 1.82 0.18 1.82 HU 1.34 1.75 0.41 1.97 MT 1.39 1.58 0.19 2.00 NL 1.72 1.82 0.10 1.78 AT 1.40 1.64 0.24 1.65 PL 1.27 1.58 0.31 2.04 PT 1.36 1.65 0.29 1.82 RO 1.32 1.75 0.43 1.93 SI 1.31 1.55 0.24 1.78 SK 1.24 1.66 0.44 2.09 FI 1.84 1.91 0.07 1.91 SE 1.85 1.96 0.11 2.00 UK 1.84 1.98 0.14 1.96 * Using the Bongaarts-Feeney formula Source: Vienna Institute of Demography, 2008 demography data sheet.

The findings analysed in this section suggest that the TFR indicator could significantly underestimate actual fertility. If confirmed, this would have major implications for future population predictions. As shown below, the potential impact of this tempo adjustment has been one of the reasons why Eurostat has assumed an increase in the TFR between 2008 and 2060 for its latest projection.

1.2. Trends in death rates

In 2005, around 4.8 million deaths were recorded in EU-27, some 300,000 fewer than the number of births. The number of deaths fluctuates much more than the number of births, thus explaining the year-on-year variations in the difference between the two figures. This difference between births and deaths is referred to as natural population change and has been positive since the mid-1990s (see Figure 1.3).

Figure 1.3: Natural population change in the EU-27 between 1996-2007

Source: Eurostat demographic data.

The total number of deaths depends, on the one hand, on the size of cohorts reaching the end of their life cycle and, on the other, on mortality rates, i.e. the likelihood of dying at a given age. One of the most impressive socio-economic achievements of developed societies has been the marked reduction in mortality or, in other words, the large increase in life expectancy. During the past 40 years, life expectancy has increased by more than 10 years for both men and women, amounting to an extra 2.5 years gained per decade. Average life expectancy at birth in EU-27 in 2004, the last year for which information is available for all 27 Member States, was 75.2 years for men and 81.5 years for women. Thus women tend to live on average 6.3 years longer than men.

Improvements in life expectancy at birth have been achieved by reducing mortality risks throughout the life cycle. A recent study carried out by NIDI

for Eurostat, covering EU-

15 countries, breaks down the changes in mortality by age groups and by main causes of death. The study found that in the 1970s life expectancy at birth increased mainly due to lower infant mortality. In the 1980s, the decline in mortality was particularly important for men in their fifties to seventies, and women over 60. In the 1990s, this trend continued, shifting to older age cohorts for both women and men (see Table 1.6); positive values in the table indicate that mortality in a particular age group has decreased, thus contributing to greater life expectancy).

18 Huisman C. (2008) Decomposition of life expectancy changes by cause of death: main findings DOC.ESTAT/F1/DEM(2008)04, study prepared for Eurostat by NIDI Netherlands. EN

Table 1.6: Arriaga decomposition of changes in life expectancy at birth by age and sex, EU-15 average, 1970-2000

Men Women

Age 1970-1980 1980-1990 1990-2000 1970-1980 1980-1990 1990-2000

0 32.7 17.3 9.4 22.4 13.7 9.1 1-4 4.4 3.7 1.0 3.0 3.5 0.8 5-9 2.2 2.2 1.1 1.0 1.4 1.1

10-14 1.5 1.3 0.9 0.6 0.9 0.8 15-19 0.6 1.9 2.2 0.7 1.0 0.6

20-24 -0.5 0.7 2.6 0.7 0.8 1.0 25-29 0.3 -1.0 2.7 1.2 0.4 1.5 30-34 2.0 -1.5 2.9 1.5 0.8 1.3 35-39 3.4 -0.6 2.5 2.2 0.8 1.6 40-44 3.9 2.1 1.8 2.9 1.9 1.1 45-49 3.5 4.8 2.6 4.3 2.4 1.8 50-54 1.7 8.0 5.1 4.5 4.2 2.7 55-59 5.7 8.5 8.6 5.0 4.8 5.1

60-64 10.6 7.6 11.2 7.0 5.2 7.9

65-69 10.2 9.7 11.5 8.6 6.8 10.2

70-74 8.2 10.3 10.4 11.7 9.1 11.3 75-79 4.4 8.6 9.9 11.4 10.4 13.6 80-84 2.9 10.4 7.4 6.9 17.5 12.9 85+ 2.3 6.0 6.2 4.4 14.2 15.5 Total 100% 100% 100% 100% 100% 100%

Change in life expectancy at birth 1.7 2.3 2.7 2.1 2.3 2.2 Source: NIDI Netherlands, "Decomposition of life expectancy changes by cause of death", 6 May 2008.

When life expectancy gains are disaggregated by main causes of death (see Table 1.7), positive values indicate that mortality from a particular cause has decreased, thus contributing to greater life expectancy. Whereas during the 1970s, much progress was achieved by reducing mortality from respiratory diseases and cancer (the former particularly in the case of men, the latter in the case of women), more recent progress was mainly due to fewer people dying from cardio-vascular diseases.

Table 1.7: Breakdown of changes in life expectancy at birth by cause of death for the EU-15 average, 1970-2000

Men 1970-1980 1980-1990 1990-2000 Infectious and parasitic diseases 10.95 2.84 -2.98

MN larynx and trachea/bronchus/lung -5.99 3.10 7.16 MN smoking related -12.11 -0.33 -0.48 MN gynaecological 0.22 0.01 -0.15

MN other and remaining 12.41 2.72 9.83

Ischemic heart disease 3.16 22.64 23.58 Other Heart diseases 15.31 11.60 11.25

Cerebrovascular disease 11.24 15.92 11.44 Respiratory system 27.29 13.97 5.23 Transport accidents 5.26 4.73 7.35 Suicide -3.02 -1.50 2.16

Remaining external causes 3.24 7.24 3.94 Alcohol related -3.51 2.87 2.63

Other remaining causes 35.54 14.18 19.04 Total 100% 100% 100%

Women 1970-1980 1980-1990 1990-2000 Infectious and parasitic diseases 6.42 2.14 -1.82

MN larynx and trachea/bronchus/lung -1.83 -0.95 -2.01 MN smoking related -5.18 -0.62 -2.33 MN gynaecological -11.45 0.06 6.97

MN other and remaining 23.47 5.68 10.80 Ischemic heart disease 5.94 12.00 18.71 Other Heart diseases 19.40 17.93 19.65

Cerebrovascular disease 13.33 21.96 20.38 Respiratory system 19.36 13.13 0.62 Transport accidents 1.75 1.40 3.57 Suicide -0.95 0.55 1.90

Remaining external causes 0.26 6.59 3.39 Alcohol related -0.68 0.95 1.75

Other remaining causes 30.15 19.17 18.42 Total 100% 100% 100% Source: NIDI Netherlands.

Not all Member States have succeeded to the same extent in reducing mortality; significant differences, therefore, remain in life expectancy at birth, as shown by the black section of the bars in Figures 1.4 and 1.5, which indicate life expectancy at birth for women and men respectively. As people reach a certain age, having survived so far, they can expect to live to a higher age than was predicted at birth.

The figures show the age up to which people can expect to live once they have reached the age of 1, 20, 40, 60, 70 and 80 years. The amount that is added to life expectancy between two ages directly reflects mortality between the ages concerned. The small increments in life expectancy (expressed as the age to which a person can expect to live) up to the age of 40 show that mortality risks have been largely eliminated during the first half of the life cycle. Thus, further gains in life expectancy will have to be achieved by reducing mortality in the second half of the life cycle.

However, there is considerable potential for improvement in several EU-12 Member States. Romania could gain about one year of life expectancy by reducing infant mortality to the EU average. As far as women are concerned, there is scope for reducing mortality between 60 and 70 in the Baltic countries and Hungary, in particular. The biggest gains in life expectancy seem possible for men aged 60 to 70, again in the Baltic countries, Bulgaria Hungary, Poland, Romania and Slovakia.

Figure 1.4: Life expectancy at different ages for women, 2004

Life expectancy at different ages, women

2004

85 70 60

80 1 0

ro lv lt pl si t gr ie nl pt at lu fi fx it bg hu ee sk cz dk uk m 27 be de cy se es eu

Figure 1.5: Life expectancy at different ages for men, 2004

Life expectancy at different ages, men

2004

r s

a 75 40 e

lv lt ro pl si pt fi lu ie at gr fx nl t it ee hu bg sk cz 27 dk be de cy uk es m se eu

Source: Eurostat demographic data.

The figures illustrate the East-West gap in life expectancy. In 2005, boys born in EU-15 could expect to live on average 6.5 years longer than their counterparts in EU-12. The gap was smaller for girls, but still amounted to 4 fewer life years for girls born in EU-12. This implies a larger gender gap in life expectancy at birth in EU-12: in 2005 it was 8.1 years, with men expecting to live 69.9 years and women 78.0 years. In EU-15, the gender gap was 5.7 years (life expectancy for men 76.4 years and 82.1 for women). The gender

gap in life expectancy at birth for individual Member States is presented in Figure 1.6. The smallest gender gaps (less than 5 years) were observed in Denmark, Greece, Ireland, the Netherlands, Sweden and the United Kingdom.

Figure 1.6: The gender gap in life expectancy at birth, in 2005

Gender difference in life expenctancy in years in 2005

difference in years

l ) d

i a i a i a r y d i a e i a i a d i n a 4 m r g y e r k s n s m n n n k r i a c n n n a g l i c 0 n c d n e l t a t v a a a n e b i u u t r i a a e a n d r u a o a a t o g l a a l a p u 0 o s l a e p d

u L n o v l g r a v S r t u l g b u r e m y M g s u m i n p 2 e r m n r l a I r e w

u P l o F o l o F o e m A e G e e C i n i t h E H S B S P m B S L R R e G D t h x K

h f r o u e d c L N

e t a i t e z a n C U ( d

l y

I t a

Source: Eurostat demographic data.

The problem that many EU-12 countries are facing is premature mortality of middle-aged men. Figure 1.7 presents the proportion of male and female deaths within the age group 40-64 in 2005 as a percentage of total deaths. Across the EU, men in this age group contribute more to the total number of deaths than women, but for the Central and East European EU-12 countries, this is particularly striking.

Figure 1.7: Proportion of deaths for men and women aged 40-64 in total deaths, in 2005

Deaths in the age group 40-64 as a percentage of total deaths

40%

35%

30%

25%

males

20%

females

15%

10%

lt lv hu pl ee sk cz ro fi bg si lu fx at dk de be pt nl es uk cy gr se

Source: Eurostat demographic data.

Another important disparity with regard to life expectancy, apart from the East-West and the gender divides, is linked to socio-economic status. A comprehensive overview of the evidence

shows that in all 10 countries examined, mortality of the lowest socio-

economic categories is 30 to 60% higher than for the highest socio-economic categories. The difference for men is larger than for women. Socio-economic status was measured by the educational attainment level, manual versus professional levels of occupation or housing quality. As shown by Table 1.8, a socio-economic gradient exists for most health indicators including life expectancy at birth. On average, less advantaged groups have shorter lives, suffer more from diseases and feel their health to be worse than more advantaged groups.

The gap in average life expectancy at birth between men from the highest and the lowest socio-economic groups is 4 to 6 years; for women, it amounts to 2 to 4 years. In some countries, these differences are considerably larger (as much as 10 years), and in many countries the gap has widened over the past three decades. People with lower education not only live shorter lives but they spend also more time in poorer health.

The issue of health inequalities is high on the agenda for the Open Method of Coordination on social protection and social inclusion. One of the common objectives agreed within this context calls for inequalities in access to care and in health outcomes to be addressed. The 2008 Joint Report on Social Protection and Social Inclusion by the European Commission and the Council

underlined that "On average, people with lower

levels of education, wealth or occupational status have shorter lives and suffer more often from disease and illness than more well-off groups and these gaps are not

19 Makenbach J. et.al. (2005) Health inequalities: Europe in Profile , Report published under the auspices of the UK presidency of the EU (October 2005). 20

See SEC(2008)91 and COM(2008)42 final EN

declining. Income inequality, poverty, unemployment, stress, poor working conditions and housing are important determinants of health inequalities, as are lifestyle and willingness and ability to bear the costs."

The Joint Report also stressed that "a combination of general policies and those tailored to lower socio-economic groups is needed " and that few Member States "have begun to address health inequalities systematically and comprehensively by reducing social differences, preventing the ensuing health differences, or addressing the poor health that results". Monitoring progress towards these goals requires further work to develop reliable and comparable indicators for life expectancy, healthy life years and infant mortality by socio-economic status.

Table 1.8: Inequalities in mortality by socio-economic status in 19 European countries

Country Indicator of Period Age- Rate Ratio Source socio- group Men Women economic

position

AT Education 1991-1992 45+ 1.43* 1.32* National census-linked mortality follow-up

BE Education Housing tenure 1991-1995 1991-1995 45+ 1.34* 1.44* 1.29* 1.43* National census-linked mortality follow-up 60-69

CZ Education End 1990s 20-64 1.66* 1.09* Unlinked cross-sectional study

DK Education Housing tenure Occupation 1991-1995 1991-1995 1981-1990 60-69 60-69 45-59 1.28* 1.64* 1.33* 1.26* 1.47* National census-linked mortality follow-up National census-linked mortality follow-up n.a.

England/ Wales Education Housing tenure Occupation 1991-1996 1991-1996 1981-1989 45+ 1.35* 1.65* 1.61* 1.22* 1.58* National census-linked mortality follow-up National census-linked mortality follow-up; representative sample 60-69 45-59

n.a.

EE Education Education 2000 1988 20+ 2.38* 1.50* 2.23* 1.31* National cross-sectional study National cross-sectional study 20-74

FI Education Housing tenure 1991-1995 1991-1995 45+ 1.33* 1.90* 1.24* 1.73* National census-linked mortality follow-up 60-69

FR Education Housing tenure Occupation 1990-1994 1990-1994 1980-1989 60-69 60-69 45-59 1.31* 1.27* 2.15* 1.14 National census-linked mortality follow-up National census-linked mortality follow-up; representative sample 1.25*

n.a.

HU Education Occupation 2002 45-64 45-64 1.97* 1.58* Cross-sectionnal ecological analysis National cross-sectional study 1984-1985 1.61 1.33

IE Occupation 1980-1982 45-59 1.38* n.a. National cross-sectional study

IT Education Housing tenure Education Occupation 1991-1996 1991-1996 1981-1982 1981-1982 45+ 1.22* 1.37* 1.85* 1.35* 1.20* 1.33* Urban census-linked mortality follow-up (Turin) National census-linked mortality follow-up National census-linked mortality follow-up 60-69 18-54 45-59

n.a. n.a.

LV Education 1988-1989 1.50 1.20 National cross-sectional study

LT Education 2001 25+ 2.40* 2.90* Unlinked cross-sectional analysis

NL Education 1991-1997 25-74 1.92* 1.28 GLOBE Longitudinal study (Eindhoven)

NO Education Housing tenure Occupation 1990-1995 1990-1995 1980-1990 45+ 1.36* 1.44* 1.47* 1.27* 1.36* National census-linked mortality follow-up National census-linked mortality follow-up 60-69 45-59

n.a.

PL Education 1988-1989 50-64 2.24 1.78 National cross-sectional study

PT Occupation 1980-1982 45-59 1.36* n.a. National cross-sectional study

SI Education 1991 & 25-64 2.44 2.66 Unlinked cross-sectional study 2002

ES Education Occupation 1992-1996 1980-1982 45+ 1.24* 1.37* 1.27* Urban and regional census-linked mortality follow-up (Barcelona & Madrid) National cross-sectional study 45-59 n.a.

SE Occupation 1980-1986 45-59 1.59* n.a. National census-linked mortality follow-up

CH Education Occupation 1991-1995 1979-1982 45+ 1.33* 1.37* 1.27* National census-linked mortality follow-up National cross-sectional study 45-59 n.a.

a Because of differences in data collection and classification, the magnitude of inequalities in health cannot always directly be compared between countries.

b Rate Ratio: ratio of mortality rate in lower socio-economic groups as compared to that in higher socio-economic groups. Asterisk (*) indicates that difference in mortality between socio-economic groups is statistically significant. N.a. indicates 'not available'.

Source: Mackenbach J. et. al. (2005) Health inequalities: Europe in profile , Erasmus University Rotterdam

1.3. Migration

In 2005, an estimated 1.6 million more people migrated to the European Union (EU-27) than from it. Migration represents a much larger contribution to current population growth than the difference between births and deaths recorded on the territories of the EU-27 Member States. Since the late 1980s, almost every year, EU-27 countries have attracted at least half a million more people than it lost due to emigration. Net migration to the EU has been particularly strong in recent years, reaching up to two million people per year in 2004 (see Figure 1.8)

Figure 1.8: Net migration (including correction) to EU-27 between 1961 and 2007

Net migration (including correction), EU27

2500000

2000000

1500000

1000000

500000

61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 05 07 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 -500000

-1000000

Source: Eurostat demographic data.

The population change due to migration is very unevenly distributed across the EU. Table 1.9 shows net migration for individual EU Member States. In recent years, Spain, Italy and the UK have received around three-quarters of the EU's migratory surplus. In Germany, which increased its population by a quarter of a million in 1996, the number was down to just below 50,000 in 2007.

21 Net migration is the difference between immigration into and emigration from the area during the year. It is therefore negative when the number of emigrants exceeds the number of immigrants. Since most countries either do not have accurate figures on immigration and emigration or have no figures at all, net migration is estimated on the basis of the difference between population change and natural increase between two dates. The statistics on net migration are therefore affected by all the statistical inaccuracies in the two components of this equation. EN

DE 398,263 281,493 93,433 46,980 202,050 167,863 274,835 EE** -15,564 -13,418 -6,927 -6,559 -1,144 224 167

IE 5,920 15,958 17,433 16,213 24,246 31,812 39,261

GR 77,285 70,975 61,409 54,818 45,016 29,401 37,779

ES 70,591 83,328 94,436 158,757 237,853 389,774 441,272 FR : : : -1,407 150,273 158,266 172,701

FX -14,567 -18,504 -13,505 -6,424 145,802 155,706 170,711 IT 28,503 56,392 50,428 55,775 34,914 49,526 49,874 CY 6,000 5,300 4,800 4,200 4,200 3,960 4,650

LV -13,713 -10,081 -9,420 -5,751 -4,085 -5,504 -5,159

LT -23,668 -23,369 -22,421 -22,122 -20,739 -20,306 -2,559 LU 4,326 3,456 3,624 3,815 4,461 3,431 3,310

HU 17,906 17,876 17,561 17,261 16,793 16,658 9,691 MT 59 264 572 353 359 : 2,173

NL 14,929 21,258 30,425 44,107 43,767 57,033 55,984 AT 2,080 3,880 1,537 8,451 19,787 17,272 43,509

PL -18,223 -12,765 -11,796 -13,261 -14,011 -409,924 -16,743 PT 21,900 25,880 28,886 31,874 38,000 47,000 65,000

RO -21,217 -19,473 -13,345 -5,629 -2,516 -3,729 -557,739 SI 777 -3,445 -1,303 -5,406 10,773 2,747 4,963 SK 2,842 2,255 1,731 1,306 1,454 -22,301 1,012 FI 4,285 3,938 4,808 4,451 3,427 2,410 6,147

SE 11,648 5,839 5,950 10,940 13,657 24,386 28,622

UK 65,026 47,867 58,407 97,371 137,647 143,871 150,956 * Migration flows have not been considered when estimating annual total population. ** Net migration including correction figures represent only corrections due to vital events. Note: High negative net migration for BG, CZ and RO in 2001 and for PL and SK in 2000 is due to Census. Source: Eurostat demographic data.

The biggest flow by far 3 million people concerns migrants from third countries migrating to the EU, most of them (2.6 million) to EU-15 countries. The EU has thus become a major destination for global migration flows, surpassing even the US.

However, migration not only occurs from third countries to the EU. The net migration figures presented in Table 1.9 mask important migratory flows within the EU. Table 1.10 presents a matrix of estimated flows in 2004 for EU-25, EU-15, EU-10 and the rest of the world. About 2.3 million people emigrated from an EU-25 country, about half to another EU-25 country and half leaving the EU. Around one-third of the intra-EU migration (0.4 million) originated from EU-10 countries with most migrants (0.3 million) moving to an EU-15 country. At the same time, 0.2 million people moved from an EU-15 to an EU-10 country.

Table 1.10: Migration flows in the EU for 2004, in millions

From EU-25 EU-15 EU-10 Non-EU Total EU-25 1.1 0.8 0.3 1.2 2.3 EU-15 0.7 0.5 0.2 0.9 1.6 EU-10 0.4 0.3 0.1 0.3 0.6 Non-EU 3.0 2.6 0.4 3.0

Total 4.1 3.3 0.7 1.2 5.3 Source: Demography Monitor 2006, NIDI Netherlands, European Observatory on Demography and the Social Situation- Demography Network, European Commission

Migrants may come to Europe for different reasons: to seek employment, rejoin family members or flee persecution or humanitarian disasters. The OECD collects data on different motives for entry, showing that, in several countries, family reasons are more important than the search for employment (see Figure 1.9).

Figure 1.9: International migration by category of entry, 2005*

International m igration by category of entry, 2005, standardised data

Employment Accompanying f amily of w orkers Family Humanitarian Other UK

0 20 40 60 80 100 % of total inf low s

Source: OECD, SOPEMI 2007.

Source: OECD (2007) International Migration Outlook SOPEMI.

Immigration to the EU has resulted in a significant proportion of third-country nationals in the EU's population. Moreover, many of those who have acquired citizenship may still be perceived as migrants or see themselves as not fully integrated in their host societies. Table 1.11 shows the proportion of non-nationals living in each Member State, also distinguishing between EU-27 citizens and non-EU citizens. At the beginning of 2007, around 4% of the EU's resident population were non-EU citizens, compared to 2.1% of EU citizens living in an EU country other than their country of citizenship. The proportion of non-EU citizens is relatively high (5% or higher) in Germany, Greece, Spain, Cyprus, Luxembourg and Austria. In the case of Estonia and Latvia, the proportion of non-EU citizens is particularly large (almost one fifth) due to the presence of so-called "recognised aliens", who have no citizenship of any existing country, Russian citizens and citizens of other countries that became independent after the end of the USSR

Table 1.11: The EU-27 population by citizenship, 1 January 2007

Total of which non-nationals As percent of the Acquisitions of citizenship population total population in 2006

non-EU-27 EU-27 non-EU- EU-27 Total as% of third- 27 country nationals

BE 10,584,534 300,816 631,345 2.8 6.0 : :

BG 7,679,290 21,690 3,800 0.3 0.0 6,738 31.1

CZ 10,287,189 186,370 109,866 1.8 1.1 2,346 1.3 DK 5,447,084 196,877 81,219 3.6 1.5 7,961 4.0

DE 82,314,906 4,788,792 2,467,157 5.8 3.0 124,566 2.6 EE 1,342,409 229,709 6,700 17.1 0.5 4,781 2.1

IE 4,312,526 141,156 311,150 3.3 7.2 5,763 4.1

GR 11,171,740 729,840 157,700 6.5 1.4 1,962 0.3

ES 44,474,631 2,856,796 1,749,678 6.4 3.9 62,375 2.2

FR 63,392,140 2,369,540 1,280,500 3.7 2.0 147,868 6.2 IT* 59,131,287 2,332,734 606,188 4.0 1.0 35,266 1.4 CY 778,684 47,184 70,900 6.1 9.1 : :

LV 2,281,305 426,687 6,264 18.7 0.3 18,964 4.4 LT 3,384,879 37,354 2,333 1.1 0.1 467 1.3

LU 476,187 27,227 170,986 5.7 35.9 1,128 4.1

HU 10,066,158 66,827 101,046 0.7 1.0 6,101 9.1 MT 407,810 4,610 9,267 1.1 2.3 474 10.3

NL 16,357,992 437,014 244,918 2.7 1.5 29,089 6.7 AT 8,298,923 550,129 275,884 6.6 3.3 25,746 4.7 PL 38,125,479 30,955 23,928 0.1 0.1 989 3.2

PT 10,599,095 339,295 95,600 3.2 0.9 3,627 1.1 RO 21,565,119 20,095 5,974 0.1 0.0 29 0.1

SI 2,010,377 50,549 3,006 2.5 0.1 3,204 6.3

SK 5,393,637 12,912 19,218 0.2 0.4 1,125 8.7 FI 5,276,955 79,277 42,462 1.5 0.8 4,433 5.6

SE 9,113,257 266,509 225,487 2.9 2.5 51,239 19.2

UK 60,852,828 2,203,028 1,456,900 3.6 2.4 154,015 7.0 *IT: Population numbers for 2006 Source: Eurostat demographic data.

In 2006, around 670 000 third-country nationals and 60 000 European Union citizens acquired the citizenship of an EU Member State. This is the same order of magnitude as for the US (703 000 in 2006)

. EU citizens living in another Member State other than their own enjoy

22 See for a more elaborate discussion of net migration and intra EU mobility the forthcoming 2008 Employment in Europe Report.

23 Source: US Department of Homeland Security, Office of Immigration Statistics, http://www.dhs.gov/xlibrary/assets/statistics/publications/natz_fr_07.pdf

most of the same rights as citizens of the Member State in which they live. Few of them may feel the need to acquire citizenship of another EU Member State.

Box 1.2: International adoptions how important are they from a demographic perspective?

It has been estimated that there have been some 20 000 international adoptions in the EU in 2004 a small number compared to the total number of births or the EU's population growth resulting from net migration. The main receiving countries in the EU seem to be Spain, France, Italy and, particularly relative to their population size, Sweden and the Netherlands.

Table 1.12: Receiving countries with the highest number of international adoptions

Country 1988 1998 2001 2003 2004 US 9,120 15,774 19,237 21,616 22,884 France 2,441 3,777 3,094 3,995 4,079 Italy 2,078 2,233 1,797 2,772 3,398

Canada 232 1 2,222 1,874 2,181 1,955 2

Spain 93 1,487 3,428 3,951 5,541 Sweden 1,074 928 1,044 1,046 1,109

Germany 875 3 922 798 674 506 Netherlands 577 825 1,122 1,154 1,307 Switzerland 492 686 457 366 557 Norway 566 643 713 714 706

Denmark 523 624 631 522 528 Belgium 662 310 255 281 470

Australia 516 245 289 278 370 Finland 78 181 218 238 289

Total 14 countries 19,327 30,801 34,870 39,696 43,699

Estimate for 20 main receiving countries n.a. 31,720 35,903 40,791 44,875 Source: Selman P. (2005), "Trends in Inter-country Adoption: Analysis of data from 20 Receiving Countries, 1998-2004", Journal of Population Research, vol. 23, No. 2/2006.

The main countries of origin of children adopted into 16 EU countries in 2003 were:

China 3,205

Russia 2,321

Colombia 1,433 Ukraine 1,234 Bulgaria 753 Ethiopia 659 Haiti 656 India 579

Vietnam 505 Brazil 439

Source: Selman P (2007), Trends in inter-country adoption 1998-2004. A demographic analysis of data from 20 receiving countries, Journal of Population Research, special issue on "Globalisation and Demographic Change"

The conclusion is therefore that international adoptions may have a major impact on the welfare of the children and the parents concerned, but their number is small. It is not a phenomenon that has a sizable demographic impact on either the sending or the receiving countries.

1.4. Europe's demographic future

This section presents the latest population projections for the European Union (EUROPOP2008 convergence scenario) and looks at some of the differences compared with the previous round of projections, carried out in 2004 (EUROPOP2004). The results of these projections are reflected in the population pyramids presented earlier in this chapter.

Demography is an area in which projections are carried out over a much longer term than in most other areas of concern to policy makers. Projections are conditional "if ..then" statements and they result of explicit assumptions that are extended far into the future. For cohorts already born, long-term projections can be quite reliable. Barring major disasters (resulting in much increased mortality or strong migratory movements), the size as well as the gender and age structure of the older cohorts can be foreseen fairly accurately. However, over a period of half a century, the projections are very sensitive to the assumptions made regarding fertility, mortality (life expectancy) and in particular migration.

For the EUROPOP2008 convergence scenario, a new conceptual framework was adopted to establish reasonable assumptions about these key determinants, taking into account the current situation, which differs widely across the EU. It would be unreasonable to assume that these differences will always prevail. The EUROPOP2008 convergence scenario is therefore based on a framework where the socio-economic and cultural differences between the Member States would fade away in the very long term. Assumptions on fertility rates for individual countries are based on a convergence trajectory whereby they would equalise across the EU by 2150. For migration, a trend towards zero net migration by 2150 is assumed. Similarly, equalisation of life expectancy across countries is assumed to take place in the very long term.

1.4.1. Fertility assumptions

The EUROPOP2008 convergence scenario is based on the expectation that fertility will rise slightly to around 1.7 children per women. By the end of the projection period (2060), significant differences in fertility are expected to persist, as shown in Figure 1.10.

Figure 1.10: Assumed Period Total Fertility Rates in 2060

2.0 2.0 1.9 1.9 1.8 1.8 EU 27 = 1.68

1.7 1.7 1.6 1.6 1.5 1.5 1.4 1.4 1.3 1.3 1.2 1.2 1.1 1.1 1.0 1.0

SK PL CZ RO SI DE HU LV LT PT BG IT MT ES EL AT CH CY EE LU N L BE FI UK DK SE IE N O FX

Source: Eurostat demographic data.

Table 1.13 presents assumed fertility rates for individual countries at 10-year intervals, showing the gradual convergence process. The assumption takes into account the possibility that fertility rates may be underestimated in countries that are currently experiencing an increase in the mean age of mothers at the birth of their first child (see section 1.1 above), but it may still underestimate rates in many EU-12 countries, where postponement could currently have a strong impact on projections. Indeed, compared to EUROPOP2004, the fertility assumptions for EU-12 have been revised downwards from 1.6 to 1.5, whereas those for EU15 have been raised by 0.1 from 1.6 to 1.7. As a result, the EUROPOP2008 convergence scenario assumes less convergence over the projection period between EU-12 and EU-15 than was the case in EUROPOP2004 (see Figure 1.11).

Table 1.13: Period Total Fertility Rate Assumptions 2008-2060

2008 2010 2020 2030 2040 2060 BE 1.75 1.76 1.76 1.77 1.78 1.79 BG 1.38 1.39 1.42 1.46 1.49 1.55 CZ 1.33 1.34 1.38 1.41 1.45 1.52 DK 1.85 1.85 1.85 1.85 1.85 1.85

DE 1.34 1.35 1.38 1.42 1.45 1.53 EE 1.55 1.55 1.57 1.60 1.62 1.66 IE 1.90 1.90 1.90 1.89 1.89 1.88 GR 1.41 1.41 1.45 1.48 1.51 1.57 ES 1.39 1.39 1.43 1.46 1.49 1.56 FX 1.98 1.98 1.97 1.96 1.95 1.93 IT 1.38 1.39 1.42 1.46 1.49 1.55 CY 1.45 1.46 1.49 1.52 1.54 1.60 LV 1.36 1.36 1.40 1.43 1.47 1.54 LT 1.35 1.35 1.39 1.43 1.47 1.54

LU 1.65 1.65 1.67 1.68 1.70 1.72 HU 1.35 1.35 1.39 1.42 1.46 1.53 MT 1.38 1.39 1.42 1.46 1.49 1.55 NL 1.72 1.72 1.73 1.74 1.75 1.77 AT 1.41 1.42 1.45 1.48 1.51 1.57 PL 1.27 1.28 1.32 1.36 1.40 1.49 PT 1.36 1.37 1.40 1.44 1.47 1.54 RO 1.32 1.33 1.37 1.41 1.44 1.52 SI 1.32 1.33 1.37 1.40 1.44 1.52 SK 1.25 1.26 1.30 1.34 1.38 1.47 FI 1.84 1.84 1.84 1.84 1.84 1.84 SE 1.85 1.85 1.85 1.85 1.85 1.85 UK 1.84 1.84 1.84 1.84 1.84 1.84 Source: Eurostat, EUROPOP2008 convergence scenario.

Figure 1.11: Total fertility rate (TFR), 1980-2006 observed, 2008-2050 projected

2,4

2,1

1,8

1,5

1,2

1980 1990 2000 2010 2020 2030 2040 2050 EU-15 obs EU-15 EP08 EU-12 EP08 EU-12 obs EU-15 EP04 EU-12 EP04

Source: NIDI based on Eurostat demographic data . 1.4.2. Life expectancy assumptions

The EUROPOP2008 convergence scenario assumes an increase in life expectancy at birth to 84.5 years for men (75.2 in 2004) and 89.0 years for women (81.5 in 2004) by the year 2060 (see Figure 1.12). Compared to 2004, this represents a gain of 9.3 years for men and 7.5 years for women. The gender gap would narrow from 6.3 to 4.5 years. The East-West gap would also be reduced.

Figure 1.12: Life expectancy at birth in 2060 for men and women

92 92

90 e° (EU27,F) = 89.0 90 88 88 86 86 e° (EU27,M ) = 84.5

84 84 82 82 80 80 78 78 76 76 74 74

BG RO LV LT HU SK EE CZ PL DK LU MT EL CY SI PT BE NL UK DE AT IE NO FI SE ES C H IT FX

The development of life expectancy over coming decades in the Member States is presented in Table 1.14. Men would typically gain about 1.5 years per decade, 2 or more years in the countries that currently have the lowest life expectancy. For women, the increments would be slightly smaller, in line with the expected reduction in the gender gap.

Table 1.14: Assumed life expectancy at birth

MEN 2008 2010 2020 2030 2040 2060 BE 76.7 77.0 78.7 80.2 81.7 84.4 BG 69.7 70.2 72.8 75.3 77.5 81.6 CZ 73.9 74.3 76.3 78.1 79.9 83.2 DK 76.4 76.8 78.4 80.0 81.5 84.3

DE 77.3 77.6 79.3 80.8 82.3 84.9 EE 68.0 68.6 71.4 74.0 76.5 80.8 IE 77.5 77.9 79.5 81.1 82.5 85.2 GR 77.4 77.8 79.4 80.9 82.3 84.8 ES 77.4 77.7 79.4 80.9 82.3 84.9 FX 77.5 77.8 79.5 81.0 82.5 85.1 IT 78.5 78.9 80.3 81.7 83.1 85.5 CY 78.2 78.5 80.0 81.5 82.8 85.2 LV 66.0 66.6 69.8 72.8 75.6 80.5 LT 65.9 66.6 69.8 72.8 75.6 80.4 LU 76.3 76.7 78.5 80.2 81.7 84.5 HU 69.7 70.2 72.9 75.4 77.7 81.9 MT 76.0 76.4 78.2 79.9 81.5 84.3 NL 77.9 78.2 79.7 81.1 82.5 84.9 AT 77.4 77.8 79.4 80.9 82.3 84.9 PL 71.4 71.9 74.3 76.6 78.8 82.5 PT 75.8 76.2 78.0 79.7 81.2 84.1 RO 69.8 70.3 73.0 75.5 77.8 81.9 SI 74.7 75.1 77.1 78.9 80.6 83.7 SK 70.9 71.4 73.8 76.0 78.2 82.0 FI 76.1 76.5 78.3 79.9 81.5 84.3 SE 79.0 79.2 80.6 81.9 83.1 85.4 UK 77.4 77.7 79.4 80.9 82.4 85.0

WOMEN 2008 2010 2020 2030 2040 2060 BE 82.3 82.6 84.0 85.4 86.6 88.9 BG 76.7 77.1 79.3 81.3 83.1 86.5 CZ 80.2 80.5 82.1 83.7 85.1 87.8 DK 81.0 81.4 83.0 84.5 85.9 88.4

DE 82.6 82.9 84.3 85.6 86.8 89.1 EE 78.7 79.1 81.1 82.9 84.5 87.5 IE 81.9 82.2 83.8 85.3 86.7 89.2 GR 82.6 82.8 84.1 85.3 86.5 87.7 ES 83.9 84.1 85.4 86.5 87.6 89.6 FX 84.3 84.6 85.8 87.0 88.1 90.1 IT 84.2 84.5 85.7 86.9 88.0 90.0 CY 81.7 82.0 83.5 84.9 86.2 88.7 LV 76.7 77.1 79.4 81.5 83.4 86.8 LT 77.4 77.9 80.0 81.9 83.7 86.9

LU 81.2 81.5 83.2 84.6 86.0 88.5 HU 78.1 78.5 80.5 82.4 84.2 87.3 MT 81.1 81.4 83.1 84.6 86.1 88.6 NL 82.2 82.5 83.9 85.3 86.6 88.9 AT 82.9 83.2 84.6 85.8 87.0 89.2 PL 79.9 80.3 82.1 83.7 85.3 88.0 PT 82.4 82.7 84.1 85.4 86.6 88.8 RO 76.6 77.1 79.3 81.3 83.2 86.6 SI 81.9 82.2 83.7 85.1 86.4 88.8 SK 78.7 79.1 81.0 82.7 84.4 87.4 FI 83.0 83.3 84.7 85.9 87.1 89.3 SE 83.1 83.4 84.7 86.0 87.2 89.3 UK 81.5 81.9 83.5 85.0 86.4 88.9 Source: Eurostat, EUROPOP2008 convergence scenario.

1.4.3. Migration assumptions

Migration patterns between the EU and the rest of the World, between Member States and within Member States can change much faster and in much less predictable ways than

fertility and life expectancy. Over recent years, the EU as a whole, and some Member States in particular, have experienced strong population growth due to net migration in excess of 1.5 million people per year since 2002. It therefore seemed reasonable to revise upwards the migration assumptions used in EUROPOP2004. Table 1.15 presents the assumptions on net migration country by country.

Table 1.15: Assumptions for annual net migration, 2008-2060

In persons 2008 2010 2020 2030 2040 2060 EU-27 1,683,921 1,563,444 1,252,781 1,093,105 1,005,488 803,507 BE 50,657 47,475 36,244 31,356 27,084 23,251 BG -1,377 419 242 -477 2,483 -1,178

CZ 24,020 25,857 24,739 22,855 27,335 16,653 DK 9,653 9,836 8,103 8,719 6,475 6,160

DE 159,773 146,680 173,142 187,050 131,599 115,852 EE -553 -392 -64 -325 75 -338

IE 63,066 53,449 21,664 8,732 6,013 8,648

GR 39,720 39,531 38,188 37,151 36,596 26,778

ES 623,449 540,207 263,065 160,787 150,488 129,859 FX 99,301 97,902 92,517 86,548 76,923 62,937

IT 259,522 255,867 240,773 248,711 229,485 174,270 CY 9,282 9,151 8,498 7,844 7,190 5,883 LV -974 -850 -333 -576 80 -592

LT -2,219 -1,706 -243 -271 -160 -113

LU 4,350 4,289 3,983 3,676 3,370 2,757

HU 19,622 19,086 22,407 17,309 22,278 14,855 MT 993 1,016 1,023 886 946 801

NL 7,846 7,736 10,618 13,668 6,508 8,350

AT 33,081 32,615 30,517 31,234 26,040 22,347

PL -15,509 -15,291 13,983 -1,340 17,059 8,154 PT 51,783 51,054 47,580 46,087 45,345 34,477 RO -5,644 -5,075 6,290 -797 12,897 3,906 SI 5,863 5,177 4,435 3,436 3,313 2,254 SK 3,552 3,210 5,001 3,866 6,081 3,682 FI 9,659 9,965 7,820 5,814 4,827 4,495

SE 46,832 42,297 26,861 20,225 17,189 15,777

UK 188,171 183,938 165,727 150,935 137,967 113,582 Source: Eurostat, EUROPOP2008 convergence scenario.

The cumulative effect of net migration assumed under the EUROPOP2008 convergence scenario is to increase the EU's population by 56 million by 2061, compared to 40 million under EUROPOP2004. This is not directly comparable because EUROPOP2004 covered only 46 years from 2004 to 2050, compared to 52 years from 2008-2060 under the EUROPOP2008 convergence scenario. As can be seen from Figure 1.13, Cyprus, Luxembourg, Spain, Portugal and Italy are expected to receive the largest net migration flows in relation to their projected total population.

Figure 1.13: Cumulative migration (2008-2060) as a percentage of the population on 1st January 2061

( % ) ( % ) 35 35 30 30 25 25 20 20 15 15 EU27 = 11.7 %

10 10 5 5 0 0

LV LT EE BG PL RO N L SK FX FI D K UK SI S E N O H U DE MT IE CZ BE AT EL CH IT P T E S LU CY

-5 -5

Source: Eurostat, EUROPOP2008 convergence scenario.

Net migration thus makes a very significant difference to Europe's future population. Without the assumed net migration inflow, Europe's population would start shrinking from 2012 onwards. With the level of migration assumed in EUROPOP2008 convergence scenario, the onset of population decline is postponed until 2035 (see Figure 1.14).

Figure 1.14: Population size of the EU-27 with and without migration, 2008-2061

P o pulation p ro jectio n s for E U27 w ith an d wi th ou t m igr atio n

s 550 n

l l i o start declin e = 2035 Mi

500

start declin e = 20 12

450

r s

400

350

300

08 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20

w ith migration without m ig ra tion Source: Eurostat, EUROPOP2008 convergence scenario.

The assumed level of immigration will also have a significant impact on the future age structure of the EU population. Table 1.16 shows that, in the absence of net migration into the EU, the old-age dependency ratio (defined as the number of people aged 65 and over divided by the number of people aged 15-64) could be almost nine percentage points higher in 2060.

For EU-12 countries, which are assumed to attract fewer migrants, the old-age dependency ratio would still be four points higher than under the zero-migration hypothesis. Thus, largescale immigration, while not preventing population ageing, does have a significant impact on the speed of ageing that the EU, and EU-15 in particular, will be experiencing over coming decades.

Table 1.16: EUROPOP2008 convergence scenario with and without migration

1-1-2008 2060 with migration 2060 without migration Difference in % and % points EU-27

Total population (x1000) 495,394 505,719 416,544 21.4 Age group 0-14 (x1000) 77,544 70,952 54,152 31.0

Age group 15-64 (x1000) 333,248 283,293 223,378 26.8 Age group 65+ (x1000) 84,602 151,474 139,014 9.0 Onset of population decline 2035 2012

Young age dependency ratio (%) 23.3 25.0 24.2 0.8 Old age dependency ratio (%) 25.4 53.5 62.2 -8.8 EU-15

Total population (x1000) 392,222 420,530 336,785 24.9 Age group 0-14 (x1000) 62,011 60,881 45,151 34.8

Age group 15-64 (x1000) 260,680 237,717 181,514 31.0 Age group 65+ (x1000) 69,531 121,932 110,120 10.7 Onset of population decline 2044 2014

Young age dependency ratio (%) 23.8 25.6 24.9 0.7 Old age dependency ratio (%) 26.7 51.3 60.7 -9.4 EU-12

Total population (x1000) 103,172 85,189 79 759 6.8 Age group 0-14 (x1000) 15,533 10,072 9 001 11.9

Age group 15-64 (x1000) 72,569 45,576 41 864 8.9 Age group 65+ (x1000) 15,071 29,541 28 895 2.2

Onset of population decline <2008 <2008

Young age dependency ratio (%) 21.4 22.1 21.5 0.6 Old age dependency ratio (%) 20.8 64.8 69.0 -4.2 Source: Eurostat, EUROP2008 convergence scenario, calculations NIDI.

*

Difference between 2060 value with migration minus 2060 value without migration in percentage of the 2060 value without migration.

1.4.4. Main results of the new projections 2008-2060

The main results of the EUROPOP2008 convergence scenario projections can be presented as a demographic balance comparing the situation on 1

January 2008 with the projected

st 24

situation for 1 January 2060 (see Table 1.17) . The results from the EUROPOP2004

baseline scenario (which used a different method for establishing assumptions for long-term trends about the main drivers of demographic change) are presented in Table 1.18. Whereas the previous projection round concluded that the population of EU-27 was likely to decline by 16 million people by the year 2050, the latest projections (EUROPOP2008 convergence scenario) expect an increase by 10 million people by the year 2060. Thus the population of EU-27 would rise from 495 to almost 506 million people. The difference is mainly due to the migration assumption, but more optimistic fertility and life expectancy assumptions for the EUROPOP2008 convergence scenario also contribute to the large difference between the two projection rounds.

24 See Eurostat Statistics in Focus 72/2008: Ageing characterises the demographic perspectives of the European societies

Table 1.17: EUROPOP2008 convergence scenario, demographic balance 1 st

January 2008 1 January 2060

(in thousands) Estimated population Cumulated births Cumulated deaths Natural change Cumulated net migration Total change Projected population 1.1.2008 2008 - 2059 1.1.2060

EU-27 495,394 250,897 298,800 -47,903 58,227 10,325 505,719 BE 10,656 6,454 6,472 -19 1,657 1,639 12,295 BG 7,642 2,739 4,941 -2,201 44 -2,158 5,485

CZ 10,346 4,364 6,433 -2,069 1,237 -832 9,514 DK 5,476 3,321 3,260 61 383 444 5,920

DE 82,179 32,206 51,693 -19,487 8,067 -11,420 70,759 EE 1,339 622 828 -206 -1 -206 1,132

IE 4,415 3,785 2,308 1,477 860 2,337 6,752

EL 11,217 4,998 6,944 -1,947 1,848 -99 1,118

ES 45,283 23,164 28,060 -4,896 11,526 6,629 51,913 FX 61,876 40,885 35,274 5,611 4,313 9,924 71,800

IT 59,529 25,453 37,412 -11,959 11,820 -140 59,390 CY 795 583 453 130 396 526 1,320

LV 2,269 871 1,453 -582 -4 -587 1,682

LT 3,365 1,331 2,145 -814 -4 -818 2,548 LU 482 353 289 64 186 250 732

HU 10,045 4,155 6,477 -2,322 994 -1,329 8,717 MT 411 187 243 -55 50 -6 405

NL 16,404 9,076 9,388 -312 504 192 16,596 AT 8,334 4,103 4,879 -776 1,480 703 9,037

PL 38,116 14,911 22,418 -7,507 530 -6,977 31,139 PT 10,617 4,938 6,603 -1,665 2,312 647 11,265

RO 21,423 8,212 13,067 -4,855 353 -4,501 16,921 SI 2,023 816 1,252 -435 191 -244 1,779

SK 5,399 2,117 3,223 -1,106 255 -851 4,547 Fl 5,260 2,999 3,227 -228 330 102 5,402

SE 9,183 5,896 5,400 496 1,196 1,692 10,875

UK 61,270 42,359 34,660 7,699 7,708 15,406 76,677 NO 4,737 3,306 2,692 614 686 1,300 6,037

CH 7,591 4,166 4,321 -155 1,757 1,602 9,193 Source: Eurostat, EUROP2008 convergence scenario.

st st

Table 1.18: EUROPOP2004, demographic balance 1 January 2004 1 January 2050, baseline scenario

(in thousands) Observed population Cumulated births Cumulated deaths Natural change Cumulated net migration Total change Population 1.1.2004 2004 - 2050 1.1.2051

EU-25 456,815 199,694 248,045 -48,351

EU-I5 382,674 170,300 207,086 -36,786 37,123 338 383,012 BE 10,396 5,022 5,427 -405 897 492 10,888

CZ 10,211 3,774 5,784 -2,010 647 -1,363 8,848 DK 5,398 2,735 3,037 -302 323 22 5,419

DE 82,532 29,880 47,191 -17,311 8,980 -8,330 74,201 EE 1,351 561 809 -248 19 -229 1,121

EL 11,041 4,352 6,559 -2,207 1,743 -464 10,578

ES 42,345 16,856 22,863 -6,007 6,235 228 42,573

FR 59,901 32,972 30,053 2,919 2,823 5,741 65,642 IE 4,028 2,718 1,903 814 645 1,459 5,487

IT 57,888 20,402 31,680 -11,278 5,777 -5,501 52,387 CY 730 401 392 8 238 247 977

LV 2,319 933 1,418 -484 30 -454 1,865

LT 3,446 1,350 1,957 -606 28 -578 2,868 LU 452 296 233 63 132 194 646

HU 10,117 4,063 6,092 -2,029 795 -1,233 8,883 MT 400 219 223 -4 113 110 510

NL 16,258 8,622 8,980 -358 1,480 1,121 17,379 AT 8,114 3,300 4,212 -912 985 73 8,187

PL 38,191 15,209 20,231 -5,022 318 -4,704 33,487 FT 10,475 4,505 5,832 -1,326 808 -518 9,957

SI 1,996 771 1,162 -390 287 -103 1,893 SK 5,380 2,111 2,892 -781 109 -671 4,709 Fl 5,220 2,573 2,875 -303 288 -15 5,205

SE 8,976 5,022 4,851 171 1,069 1,240 10,216

UK 59,652 31,047 31,390 -343 4,939 4,596 64,247 BG 7,801 2,229 4,740 -2,512 -252 -2,764 5,038

RO 21,711 7,947 12,194 -4,247 -475 -4,722 16,989 Source: Eurostat, EUROP2004.

Striking differences are also found between the two projection rounds for individual countries. Italy's population, according to the EUROPOP2008 convergence scenario, would be the same size as today whereas, under EUROPOP2004, it was expected to fall by 5.5 million. The UK is expected to become the most populous EU-27 country with almost 77 million inhabitants in 2060, according to the EUROPOP2008 convergence scenario. The previous projection round expected a population of 64 million UK residents in 2050.

These differences between the two rounds of projections underline the importance of interpreting such results with caution. Nevertheless, one development is certain, namely the forthcoming retirement of the baby-boom cohorts which will accelerate the process of population ageing and shift the balance between people of working age and retirees (see Chapter 3). Both projection rounds also yield very similar results as far as the long-term evolution of the old-age dependency ratio (people aged 65+ in relation to people aged 15-64) is concerned: EUROPOP2004 expected a ratio of 0.53 for 2050 (EU-25); according to the EUROPOP2008 convergence scenario, it should be 0.50, rising to 0.53 in 2060. Today's oldage dependency ratio is 0.25, meaning that for every person aged 65 or over, there are four people of working age (15-64). In 2050, there will be only two people of working age for every person aged 65 or over (see Table 1.19). Only unrealistically large increases in net immigration or birth rates could curb this trend to a noticeable extent, and this would imply very rapid population growth. By contrast, a shrinking population resulting from very low birth rates and an unfavourable net migration balance can seriously accelerate the ageing of a

country's population. Countries in this situation can expect their old-age dependency ratios to triple compared to today, rising to levels as high as 0.68 in Slovakia and Poland, for two people aged 65 or over, there would only be three of working age.

Table 1.19: Old-age dependency ratios for selected years, 2008-2060

(%) 2008 2010 2020 2030 2040 2050 2060

EU 27 25.39 25.90 31.05 38.04 45.36 50.42 53.47 BE 25.80 26.09 30.60 37.58 42.27 43.87 45.84 BG 24.99 25.29 31.10 36.28 43.58 55.44 63.54 CZ 20.59 21.83 31.07 35.71 42.71 54.81 61.40 DK 23.61 24.98 31.85 37.85 42.69 41.31 42.66 DE 30.29 31.17 35.28 46.23 54.73 56.43 59.08 EE 25.23 25.01 29.18 34.42 38.96 47.19 55.55 IE 16.31 16.67 20.23 24.63 30.60 40.40 43.57 EL 27.77 28.22 32.75 38.47 48.25 56.99 57.12 ES 24.15 24.43 27.42 34.32 46.39 58.69 59.07 FX 25.33 25.81 32.77 39.02 43.99 44.68 45.20 IT 30.47 30.99 35.47 42.45 54.07 59.24 59.32 CY 17.69 18.00 22.26 27.44 30.76 37.65 44.47

LV 25.02 25.17 28.08 34.57 40.72 51.18 64.45 LT 23.02 23.18 25.98 34.71 42.81 51.13 65.65 LU 20.92 21.07 24.23 30.80 36.31 37.82 39.10 HU 23.50 24.22 30.31 34.06 40.11 50.83 57.64 MT 19.79 21.19 31.25 39.14 41.71 49.77 59.07 NL 21.84 22.82 30.69 40.00 46.77 45.61 47.18 AT 25.43 26.01 29.18 38.09 46.03 48.31 50.65 PL 18.95 18.98 27.19 35.98 41.29 55.69 68.97 PT 25.91 26.58 30.66 36.63 44.59 52.96 54.76 RO 21.34 21.34 25.67 30.32 40.75 54.00 65.27 SI 22.97 23.91 31.21 40.83 49.40 59.40 62.19 SK 16.58 16.95 23.85 32.30 39.98 55.46 68.49 Fl 24.80 25.70 36.75 43.89 45.06 46.61 49.30 SE 26.66 27.81 33.69 37.43 40.78 41.91 46.71 UK 24.27 24.72 28.58 33.23 36.92 37.96 42.14 NO 22.10 22.73 28.32 34.32 40.24 41.43 43.92 CH 24.10 24.94 29.93 37.72 43.74 45.74 48.51 Source: Eurostat, EUROPOP2008 convergence scenario .

The trend towards a very different age structure of the population is inevitable. The proportion of people of working age, aged 15-64, will decline while the proportion of older people aged 65 or over will increase. The fastest growing age group, however will be people aged 80 or over whose proportion in the population could almost triple from 4% to 11% (see Table 1.20).

Table 1.20: Projected age structure changes in the EU between 2008 and 2050

Shares in % 0-19 20-64 65-79 80+ population in millions

2008 2050 2008 2050 2008 2050 2008 2050 2008 2050 EU-27 22 19 61 52 13 18 4 11 495.4 515.3 BE 23 21 60 53 12 16 5 10 10.7 12.2 BG 20 17 63 52 14 22 4 10 7.6 5.9 CZ 21 17 65 52 11 22 3 9 10.3 9.9 DK 25 22 60 54 11 15 4 10 5.5 5.9

DE 19 16 61 52 15 18 5 14 82.2 74.5 EE 22 19 61 53 14 18 4 9 1.3 1.2 IE 27 23 62 53 8 16 3 7 4.4 6.5

GR 19 18 62 51 15 20 4 11 11.2 11.4 ES 20 18 64 50 12 21 5 11 45.3 53.2 FX 25 23 59 52 11 15 5 10 61.9 71.0 IT 19 17 61 51 15 20 5 13 59.5 61.2 CY 25 20 63 57 10 16 3 7 0.8 1.3

LV 21 17 62 54 14 20 4 10 2.3 1.8 LT 23 16 61 54 13 19 3 11 3.4 2.7 LU 24 22 62 55 11 14 3 9 0.5 0.7

HU 21 17 63 53 12 20 4 9 10.0 9.1 MT 23 17 63 54 11 19 3 10 0.4 0.4

NL 24 20 61 53 11 16 4 11 16.4 16.9 AT 21 18 62 53 13 17 5 11 8.3 9.1

PL 23 16 64 53 10 22 3 10 38.1 33.3 PT 21 18 62 52 13 20 4 10 10.6 11.4 RO 22 16 63 53 12 22 3 9 21.4 18.1 SI 20 17 64 50 13 21 4 12 2.0 1.9 SK 23 15 65 53 9 22 3 9 5.4 4.9

FI 23 21 60 52 12 16 4 11 5.3 5.4

SE 24 22 59 54 12 15 5 10 9.2 10.7 UK 24 22 60 55 12 14 5 9 61.3 74.5 Source: Eurostat, EUROPOP2008 convergence scenario. 1.5. Ageing as a global phenomenon

Ageing does not affect only EU countries, although, together with Japan, many EU Member States are among the most aged countries in the world. Population ageing is a universal process that accompanies economic and social development. Indeed, the pace of population ageing can be faster in developing than in developed countries, requiring them to adjust to the rapidly growing number of older people.

The United ations ageing index illustrates the pace of the ageing process across the world (see Table 1.21). It presents the number of older people (defined here as people aged 60 or more) per 100 younger people (aged 0-14 years). An increase in this index means that the population is ageing. In 2007 the index stood at 136.2 for Europe (defined more broadly than the EU) compared to 38.7 for the world as a whole. For the developed regions of the world, the index was 124.2. Within Europe, the South and West are most aged (indices of 155.6 and 147.3 respectively). By 2050, the ageing index is expected to less than double in Europe, but more than triple in the less developed regions, which, by then, will have a similar proportion of people over the age of 60 to that in Europe today. Thus, Europe will not be alone in having to tackle the challenge of an ageing population.

Table 1.21: UN Ageing indicators for world regions in 2007 and 2050

Ageing index % Population Median age Dependency ratio

2007 100 * b/a a: 0-14 b: 60+ Youth Old World 38.7 27.6 10.7 28.1 42.4 11.5

More developed regions 124.2 16.7 20.7 38.6 24.6 22.9 Less developed regions 28.0 30.0 8.4 25.6 46.7 8.8

Least developed regions 12.4 41.3 5.1 18.9 74.5 5.9

Africa 12.9 41.1 5.3 18.9 74.0 6.2 Asia 35.8 27.0 9.6 27.7 40.6 9.9

Europe 136.2 15.5 21.1 39.0 22.6 23.5

Latin America and the Caribbean 31.3 29.2 9.1 25.9 45.3 9.8 Northern America 86.1 20.1 17.3 36.3 29.8 18.6 Oceania 59.7 24.2 14.4 32.3 36.9 15.6

Eastern Europe 123.4 14.9 18.3 37.5 20.9 19.8

Northern Europe 124.3 17.5 21.7 38.9 26.2 23.9 Southern Europe 155.6 14.9 23.2 39.8 22.2 26.5 Western Europe 147.3 15.9 23.4 40.7 24.0 26.9 Northern Africa 21.4 32.3 6.9 23.0 51.3 7.4 Western Asia 20.1 33.1 6.7 23.6 53.2 7.3 2050

World 107.4 20.2 21.7 37.8 31.7 25.4

More developed regions 206.8 15.6 32.4 45.5 26.8 44.4 Less developed regions 95.7 20.9 20.0 36.6 32.4 22.6 Least developed regions 34.1 28.9 9.9 27.3 44.9 10.2

Africa 34.7 28.7 10.0 27.4 44.4 10.3 Asia 129.1 18.3 23.6 39.9 28.5 27.2

Europe 229.7 15.0 34.5 47.1 26.2 48.0

Latin America and the Caribbean 133.4 18.1 24.1 39.9 28.4 28.9 Northern America 157.6 17.1 27.0 41.5 27.7 34.2 Oceania 139.0 18.0 25.0 40.5 28.7 30.8

Eastern Europe 230.6 14.9 34.5 47.2 25.4 44.4

Northern Europe 187.0 16.2 30.2 43.7 27.0 40.0 Southern Europe 276.3 14.0 38.6 50.1 26.1 60.7 Western Europe 222.5 15.2 33.9 46.6 26.7 48.5 Northern Africa 94.3 20.6 19.4 36.1 31.5 21.2 Western Asia 84.2 21.2 17.8 35.3 32.1 19.4 Source: UN (2007), the 2006 revision of the Medium variant of the 2004 UN World Population Prospects.

2.

C HANGING F AMILY AND H OUSEHOLD L IVING AND W ORKING A RRANGEMENTS This chapter examines how the demographic and socio-economic trends reported in Europe in the latter part of the 20

th th

century and early 20 century are affecting family life. It begins by

presenting the changing definitions of families and households used by demographers, before exploring the impact on family composition of the trends analysed in Chapter 1. It then looks at the ways in which demographic change shapes household size and structure. The chapter concludes with a review of the challenges that changing patterns of family formation and dissolution, living and working arrangements present for governments and comments on their implications for policy.

25 th

The first demographic transition , which continued throughout most of the 20 century in the

Western world, was characterised by decreasing fertility and mortality and increasing life expectancy. The second demographic transition began in the 1960s at the time when the postwar baby boom was coming to an end, and fertility rates were falling below the replacement level. This resulted in the low population growth and population ageing described in Chapter 1.

Another significant change was the increase in divorce rates. By the 1970s, alternatives to formal marriage were developing. In some countries, the number of couples cohabiting before marriage and the number of extramarital births were increasing rapidly.

A further trend contributing to changing family life was the dramatic increase in women's educational attainment and labour market participation from the 1970s in all European countries. These combined trends resulted in a diversification of family living arrangements, smaller family and household size, a growing number of lone-parent families and singleperson households.

This de-institutionalisation of family life is closely interconnected with the processes of slower population growth and accelerated population ageing. The reduction in the time devoted to childbearing and childraising associated with the postponement and decline in fertility have made women less dependent on the bonds of formal marriage for their livelihood. In turn, lower levels of long-term commitment to marriage and its instability may be linked to the fall in fertility rates and smaller family size. Greater life expectancy, population ageing and increasing geographical mobility have called into question the relationships between the generations and the availability of family support networks. The diversification of family forms and structures has created important policy dilemmas for governments concerned about the threat posed by family breakdown for social order and the social and economic well-being of family members throughout the many stages in their life course.

2.1. Defining families and households

Family is a shifting concept. What it means to be a member of a family and the expectations people have of family relationships vary over time and space, making it difficult to find a universally agreed and applied definition. In their attempt to capture and track changing

25 See chapter 2 of the Commission's first demography report (Europe's demographic future: facts and figures on challenges and opportunities , SEC(2007)638).

family forms and composition, demographers most often refer to the family nucleus and to private household units. Due to differences in the timing and formal recognition of changing patterns of family formation and dissolution, these concepts have become more difficult to operationalise. Analysts of demographic statistics therefore have access to relatively few complete and reliable datasets with which to make comparisons over time and between and within countries.

To assist governments in collecting data that can be collated internationally, the United Nations Economic Commission for Europe (UNECE) in conjunction with Eurostat draws up recommendations for the definitions to be used in the national censuses that are usually carried out every 10 years in the Member States of the European Union. Box 2.1 presents the definitions of family and household recommended for the forthcoming 2010 round of censuses.

Box 2.1: Defining the family nucleus and household

In family statistics, the term family nucleus is often used to refer to two or more persons who live in the same household and who are related as husband and wife, as opposite-sex partners in a registered partnership or as cohabiting partners, as a marital (registered) same-sex couple, or as parent and child.

When a family is defined in this narrow sense it may consist of a couple without children, a couple with one or more children or a lone parent with one or more children.

A private household, by contrast, is either a housekeeping unit, consisting of one or several persons occupying the whole or part of a housing unit, who provide themselves with the essentials for living, or a dwelling unit with one or more persons living in a housing unit (particularly relevant in the case of register-based statistics).

The concept of a private household thus applies to both:

-

non-family households when one person lives alone or when two or more persons living in the same household unit do not constitute a family nucleus; -

and family households when two or more persons do constitute a family nucleus (a

family household can also consist of two or more families).

The United Nations' definitions have been modified since they were first established in the 1970s to take account of changes in family and household composition. For example, the definition of the family nucleus was originally based on the `conjugal family concept' according to which married couples were counted as family nuclei whether or not they had children. As consensual unions have been more widely recognised in national statistics, definitions have been extended: the recommendations for the 2010 censuses include registered same-sex couples in the list of related persons who can be considered to form a family nucleus.

Source: United Nations Economic Commission for Europe/Statistical Office of the European Communities (Eurostat) (2006) Conference of European Statisticians: recommendations for the 2010 Censuses of Population and Housing , prepared by the United Nations Economic Commission for Europe and Eurostat, New York/Geneva.

Although there is considerable overlap between the concepts of household and family in the statistics, the two concepts are interchangeable. Not every household can automatically be regarded as a family, and not every family forms a simple household. In addition, family ties

often extend beyond households as generations no longer cohabit, and family units break down or reconstitute. In this chapter, the term family is based on the concept of the family nucleus and, to some extent, family ties reaching beyond the family nucleus. The term household is used to encompass all living arrangements including one-person households.

2.2. Trends in family formation and composition

Most of the key changes that demographers were identifying in family formation and composition in European societies during the latter part of the 20

century have continued

into the 2000s. In some cases, they have been intensified, and in others they appear to be stabilising, insofar as can be assessed from the available data. This section examines in more detail how the key components of family formation and dissolution evolved in the late 20

and early 21 centuries, and tracks the resulting diversification of family forms.

2.2.1. Changing patterns of family formation, dissolution and reconstitution

For national statistical offices, registered marriage has long been considered to mark the first stage in family formation. In their efforts to make well-informed policy, governments have collected data on marital status (single, married, widowed or divorced and not remarried), gross (crude) marriage rates, age-specific first marriage rates, mean age at first marriage and at all marriages, proportions of first-married men and women by generation, divorce rates by duration of marriage and median duration of marriage at divorce. As alternatives to legal marriage, as well as divorce have become more widespread, data have also been collected on non-marital living arrangements. This section examines the different stages in couple formation, dissolution and reconstitution through marriage and divorce, unmarried cohabitation and separation, remarriage and re-partnering where the available data allow comparisons to be made over time and across countries.

2.2.1.1. Changing patterns of marriage and divorce

According to Eurostat, the number of all marriages contracted in a given year in EU-27 between 1975 and 2005 declined by almost 30%. In 1975, 3.45 million marriages were recorded, but by 2005 the number had fallen to 2.4 million. One reason for this decline has been the ageing of the European population, which led to a decrease in the number of young people, automatically resulting in fewer new candidates for marriage.

To find out whether the preference for first marriage has indeed decreased over the last decades, a correction has to be made to account for the change in the total number of potential marriage candidates. Table 2.1 presents the total rate of first marriages adjusted for changes cohort size

. It confirms that the rate of first marriage has fallen everywhere in EU-27. In

1975, more than about 80% of all men and women, with the exception of the Nordic countries, entered into a first marriage. By 2003, the rate had fallen to about 55%. More recent information from the 1990s and the beginning of 21

century for other European

countries seem to confirm the general decline in the rate of first marriage, although Denmark, France, Romania and Sweden were reporting a slight recovery. The rate of first marriage for women is higher than for men because men are more likely than women to enter into a second marriage.

26 But even after this correction the Total First Marriage Rate may still be biased. The TFMR is (like the TFR) a period estimator and as such sensitive to a postponement in the age of first marriage.

Table 2.1: Trends in total rate of first marriages, in %*

1975 1980 1985 1990 1995 2000 2003

Men

EU-25 86 74 67 67 54 57 53 BE 86 74 62 67 52 48 44 BG 94 90 87 83 54 50 46 CZ : : : : : 48 43

DK 62 49 54 56 61 69 68 DE 77 68 61 59 49 52 50 EE : : : : : 37 41

IE : : 70 71 60 : :

EL : : : 72 73 50 62

ES : 79 64 67 58 59 54

FX** 82 69 53 55 48 58 55 IT 91 79 69 69 59 58 : CY : : : : : : :

LV : : : : : 39 44 LT : : : : : 54 55

LU : : : 58 50 49 44

HU 94 77 80 77 57 48 44 MT : : : : : 87 72

NL 78 66 55 62 49 54 52 AT 73 67 59 54 50 50 48 PL : : : : : 63 58

PT : : 79 87 76 72 59 RO : : : : : 60 64

SI 94 73 61 49 49 42 40 SK : : : : : 51 50

FI 64 61 55 54 52 59 59 SE 57 49 49 52 42 49 46 UK : 76 65 59 50 49 :

Women EU-25 89 77 65 72 57 51 47 BE 89 77 65 72 57 51 47 BG : 97 93 87 56 53 49

CZ : 78 122 103 : 50 45 DK 67 53 57 60 65 73 71 DE 80 69 63 64 56 59 55 EE 94 94 88 79 45 37 42 IE 94 75 69 70 59 : : EL : : : 72 75 54 68

ES : 76 64 69 60 63 58

FX** 86 71 54 56 49 61 57 IT 94 78 68 69 62 64 : CY : : : : : : :

LV : 97 93 94 47 40 45

LT : 94 98 106 67 56 56 LU : : : 64 56 54 50

HU : 89 86 77 56 49 47 MT : : : : : 89 76

NL 83 68 57 66 53 59 55 AT 74 66 60 58 55 55 51 PL 92 89 89 90 66 63 58 PT : 86 79 88 77 75 63

RO 98 102 85 94 73 64 69 SI 99 79 65 51 51 45 42 SK : : : : : 52 52

FI 70 67 58 59 57 62 62 SE 63 52 53 55 44 53 52 UK : : 66 62 53 54 :

*

The mean number of first marriages per woman or man in a given year adjusted for cohort size. First marriages by age group are added together assuming that the number of men and women in each group is the same. This is the first marriage rate of a hypothetical individual subjected at each age to the current marriage conditions. It is comparable in nature to the period total fertility rate (see chapter 2). ** Metropolitan France

Source: NIDI and Eurostat demographic data 2006, corrections and changes have been introduced by the authors.

A number of reasons have been proposed for the decrease in first marriages across the EU in the late 20

century. The spread of reliable methods of birth control in Western societies and

the rise in female paid employment made women less dependent on a formal marital relationship for their livelihood. As more emphasis was placed on individual self-fulfilment, couples may have become less willing to accept the kind of compromise needed to support a marital partnership. Traditional marriage was no longer the only option. Changes in family law making divorce easier and establishing alternative contractual living arrangements, together with the decline in religious observance, endorsed and reinforced changing value systems and social behaviour.

Like the age of the mother at first birth (see Chapter 1), the average age at first marriage also increased markedly between 1990 and 2003: by 2.3 years for men and by 2.6 years for women (see Table 2.2). The age gap between men and women narrowed from 2.7 years in 1990 to 2.4 years in 2003. The gap between the countries with the highest and lowest age at first marriage was reduced slightly from 6.5 years in 1990 to 6.3 years in 2003. As with childbirth, women in the Central and Eastern European countries continued to enter into a marriage at a younger age than in EU-15 countries. No clear link can be observed between the marriage rate and mean age at first marriage. Even in countries with a relatively high mean age of first marriage, the marriage rate may be relatively high. A positive correlation is found, by contrast, between postponement of marriage and postponement of first birth.

Table 2.2: Change in average age at first marriage between 1990 and 2003

1990 2003

Women Men Women Men EU-25* 24.8 27.5 27.4 29.8 BE 24.3 26.3 27.1 29.3 BG 21.4 24.6 24.9 28.2 CZ 21.1 23.5 25.6 28.4 DK 27.6 30.0 30.1 32.3 DE 25.3 27.9 28.1 30.6 EE 22.5 24.6 25.5 28.1 IE 26.5 28.3 : :

EL 24.7 28.7 27.3 31.0 ES 25.3 27.5 28.3 30.2 FR 25.6 27.5 28.2 30.4 IT 25.6 28.6 27.4 30.4 CY : : 26.1 28.9

LV 22.3 24.1 24.8 26.8 LT 22.4 24.2 24.4 26.6 LU 25.4 26.9 27.9 30.2 HU 21.5 24.2 25.8 28.6 MT : : 26.5 29.0

NL 25.9 28.2 28.4 30.8 AT 24.9 27.4 27.4 29.9 PL 22.7 : 24.7 27.0

PT 23.9 26.0 26.1 28.0 RO 22.0 : 24.1 27.5

SI 23.8 26.6 27.5 30.1 SK 21.8 24.8 25.0 27.7 FI 25.0 27.0 28.5 30.4 SE 27.5 29.9 30.5 32.9 UK 25.0 27.2 27.2 29.3 *estimate Source: Eurostat .

Another important phenomenon in the EU is the number of mixed marriages between spouses of different EU nationalities, as well as between EU nationals and immigrants from third countries. Mixed marriages are defined as marriages where one of the spouses is a citizen of the county in question whereas the other is not. Figure 2.1 presents the situation across the EU in 2006 (or last year available)

. The number of mixed marriages was high more than 20%

of all marriages in Cyprus, Luxemburg and Estonia, followed by Austria and Belgium. It was lowest, at around 5%, in Hungary, Romania, Latvia and Finland. Currently, mixed marriages between EU nationals represent an important proportion of total marriages in most EU Member States.

Figure 2.1: Proportion of mixed marriages among all marriages concluded in 2006

0 , 3 5

0 , 3 0

0 , 2 5

0 , 2 0

0 , 1 5

0 , 1 0

0 , 0 5

0 , 0 0

B E B G C Z D K D E E E E S F R I T C Y L V L T L U H U M T N L A T P T R O S K F I S E H R T R S h a r e o f m i x e d m a r r i a g e s i n t o t a l m a r r i a g e s

Source: IHS Austria based on national statistical data.

If marriage marks the institutionalisation of family formation, divorce formally registers the end of a relationship. Divorce rates fluctuate with changes in the law and, thus, cannot provide reliable data on trends over time and space in the rate of marital breakdown. A steep increase or fall in a given year may be explained by the introduction of more permissive or restrictive regulations, thereby masking longer-term trends. In most countries, legislation in the late 20

century facilitated access to divorce by mutual consent, in general contributing to

an increase in divorce rates.

Table 2.3 presents the mean number of divorces in a given year in relation to the number of marriages concluded in that year (corrected for differences in the size of marriage cohort). As with total fertility and total marriage rates, the total annual divorce rate is not the divorce rate of an actual 'marriage cohort'; rather, it is the divorce rate of a hypothetical generation subjected at each age to current marriage and divorce conditions, unbiased by the age structure of the population, thereby ensuring greater comparability over time and across countries.

27 Schuh U. (2008) Mixed marriages in the EU Research Note HIS Austria, European Observatory on Demography and the Social Situation- Demography Network, European Commission.

Table 2.3: Trends in divorce rates*, 1975-2005

1975 1980 1985 1990 1995 2000 2005 BE 0.2 0.2 0.3 0.3 0.5 0.4 : BG : : : : 0.2 0.2 0.3 CZ : : : : 0.4 0.4 0.5

DK 0.4 0.4 0.5 0.4 0.4 0.4 0.5 DE 0.3 0.3 0.3 0.3 0.3 0.4 0.4 EE : : : : 0.7 0.5 :

IE 0.0 0.0 0.0 0.0 0.0 : :

GR 0.1 0.1 0.1 0.1 0.2 0.2 :

ES 0.0 0.0 0.1 0.1 0.1 0.2 0.3 FR** : : : : : : :

IT 0.0 0.0 0.0 0.1 0.1 : 0.1 CY : : : : 0.1 0.2 0.2 LV : : : : 0.3 0.3 0.4 LT : : : : 0.3 0.4 0.5

LU 0.1 0.3 0.3 0.4 0.3 0.5 0.5 HU : : : : 0.3 0.4 0.4 MT** - - - - - - -

NL 0.2 0.2 0.3 0.3 0.4 0.4 0.4 AT 0.2 0.3 0.3 0.3 0.4 0.4 0.5 PL : : : : 0.1 0.2 0.3

PT 0.0 0.1 0.1 0.1 0.2 0.3 0.3 RO : : : : 0.2 0.2 0.2

SI : : : 0.1 0.1 0.2 0.3 SK : : : : : 0.3 0.4

FI 0.3 0.3 0.3 0.4 0.5 0.5 0.5 SE 0.5 0.4 0.5 0.4 0.5 0.5 0.5 UK 0.3 0.3 0.4 0.4 0.4 : 0.4 * Independent of marriage cohort size ** Data for France are missing; divorce is not permitted in Malta. Source: Eurostat demographic data .

The data presented in Table 2.3 indicate a steady increase in divorce rates in most Member States. Whereas levels were already relatively high in some countries in the 1970s, most notably Denmark and Sweden, no statistics on divorce were collected in Ireland, Italy, Spain and Portugal in 1975. No data are available for the earlier period in the countries that joined the EU in 2004 and 2007, but recent data show large disparities in their divorce rates in the 21

century: Estonia and Lithuania record high rates, similar to those in the Nordic states,

whereas Bulgaria, Poland, Slovenia and Romania are closer to the countries in Southern Europe, displaying some of the lowest rates. Between 2000 and 2005, divorce rates appear to have stabilised in a number of countries, including Finland and Sweden where unprecedented levels had been reached in 1995, with 1 divorce for every 2 marriages. The disparity between the countries with the highest and lowest rates has changed very little over the period.

Whereas the number of divorces was increasing during the late 20 century, the mean 28

duration of marriage at the time of divorce generally did not fall over the period. Table 2.4

shows that, in most of the countries with complete datasets, couples were divorcing after a larger number of years of marriage in 2005 than in 1975, suggesting that marriage continues to be seen as an enduring institution. The disparity between the countries with the longest and shortest duration of marriage at the time of divorce decreased by the end of the period,

28 The mean duration of marriage at divorce is obtained by adding the series of divorce rates by duration of marriage for the calendar year under consideration, and calculating the mean of this sum.

ranging from 8.4 in Austria to 24.2 years in Italy in 1975, but from 10.5 in Latvia to 16.8 in Italy in 2005.

Table 2.4: Trends in mean duration of marriage (in years) at the time of divorce, 1975-2000

1975 1980 1985 1990 1995 2000 2005 BE 13.3 12.6 13.6 13.9 13.3 13.1 : BG : : : : 9.0 10.2 12.3

CZ : : : : 10.7 11.2 12.3

DK 11.1 10.4 11.2 11.4 11.4 11.5 11.4 DE 9.1 10.0 10.3 11.4 11.7 12.3 12.7 EE : : : : 10.1 9.8 :

GR 11.5 14.7 12.8 12.1 11.1 12.4 : ES : : 15.4 14.8 16.1 14.7 13.9 FR* : : : : : : :

IT 24.2 17.1 16.9 15.5 15.8 : 16.8 CY : : : : 11.4 12.0 11.9 LV : : : : 9.9 10.4 10.5

LT : : : : 11.0 11.4 11.6

LU 10.5 11.4 11.5 13.1 12.5 11.5 13.3 HU : : : : 10.8 11.0 11.8

NL 14.4 11.5 12.1 11.8 11.5 12.2 13.0 AT 8.4 9.2 9.8 10.0 10.7 11.1 10.7 PL : : : : 12.5 11.4 13.3

PT 17.4 13.9 14.3 14.2 13.5 13.3 12.7 RO : : : : 9.1 9.9 11.1

SI : : : 10.4 12.2 12.6 13.6 SK : : : : : 11.8 13.0

FI 11.6 11.6 12.4 12.4 12.5 12.7 12.6 SE 12.6 11.2 11.8 12.0 11.8 11.7 11.9 UK 13.0 12.5 11.1 11.6 11.2 : 13.3 * Data for France are missing Source: Eurostat demographic data .

2.2.1.2. Trends in unmarried cohabitation, separation and re-partnering

Legally contracted marriage and divorce provide only a partial view of changing patterns of couple formation and dissolution. A marked trend since the 1980s, firstly in the Nordic countries and subsequently extending southwards, is the development of unmarried cohabitation. Initially, consensual unions served as a prelude to marriage to the extent that unmarried cohabitation replaced marriage as the first form of partnership for young people. By the 1990s, consensual unions had come to be seen as a longer-term alternative to marriage in these countries. Today, young people at the beginning of their working life and well into their twenties are likely to be spending more time in education and training and may embark on one or more cohabiting relationships without necessarily seeing them as long term or as a prelude to raising a child.

Unmarried cohabitation is difficult to define and to measure. Unlike officially recorded life events such as births, marriages and deaths, consensual unions are generally not registered with administrative services. However, as unmarried cohabitation has become more widespread, some governments have given legal status to non-marital relationships, recognising the rights of unmarried heterosexual and same-sex partnerships. As a result, nonmarital partnerships have become institutionally visible and can be recorded in official statistics. The data from registers are, however, usually confined to crude rates by age and sex with little information about duration or separation. Where consensual unions are not legally registered, either because they have no legal status, or the partners choose not to register, the data collected are based on self-reporting.

The 2001 population census provides information about the proportion of couples living together without being formally married, but no similar EU-wide data are available on samesex cohabiting couples or couples living together for only part of the time. Table 2.5 shows how unmarried cohabitation has become a widespread and socially acceptable living arrangement in different regions within the EU. On average, about 9% of all couples are cohabiting, but large differences are found between countries. The proportion of cohabiting couples is largest in Northern Europe (22% for Denmark and Finland and 21% in Estonia) and much lower in Southern and Eastern Europe.

Table 2.5: Proportion of unmarried cohabiting couples in 2001 by age groups, in %

total 20-29 30-39 40-49 50+ EU23* 9 31 13 7 3 BE 9 35 14 7 3 BG 8 26 8 5 2 CZ 5 13 6 5 4

DK 22 69 30 16 8 DE 10 40 16 8 4

EE 21 52 26 17 11 IE 10 52 13 5 2 GR 3 12 4 2 1 ES 6 25 9 5 2 IT 4 12 6 4 2 Cy 1 6 2 1 0

LV 9 20 9 8 7 LT 7 13 8 7 5

HU 11 30 13 10 6 NL 16 57 22 10 5 AT 12 40 17 9 5 PL 2 6 2 2 1

PT 7 15 9 7 4 RO 8 18 8 6 4

SI 9 32 16 8 3 SK 3 4 3 2 2

FI 22 65 30 18 8 UK 16 55 23 12 5 * No information available for FR, LU, MT and SE. Source: Eurostat 2001 Population Census .

As divorce has become more widespread in Europe, a growing number of marriages are remarriages involving divorced persons. Table 2.6, which presents data for 1995 to 2005, shows that, although most people marrying in EU-27 were single (rather than divorced or widowed), a substantial minority of marriages were of divorcees. The Southern European countries and Poland and Slovenia, which recorded relatively low divorce rates, reported the lowest proportion of remarriages. In Austria, Germany, Luxembourg, the Czech Republic, Denmark, Estonia, Hungary and Latvia, by contrast more than 20% of all marriages in 2005 involved persons previously married. Except for the Southern European countries, where the proportion of divorcees in all marriages has remained relatively stable, in most countries re-partnering through marriage has become widespread.

Table 2.6: Proportion in all marriages of previously divorced persons, 19952005, in %

men women

1995 2000 2005 1995 2000 2005 BE 21 23 : 21 23 :

BG 11 12 12 9 11 11

CZ 23 24 25 23 24 24 DK 22 21 22 22 21 21 DE 21 24 25 23 26 27 EE 31 29 29 28 29 25 IE 0 : 7 0 : 5

GR 9 11 11 8 10 11 ES 5 6 8 4 5 8

FX* 16 17 19 16 16 18 IT : 6 7 : 5 7

CY : 21 16 : 19 15 LV : 27 26 : 25 22

LT 17 20 22 14 18 19 LU 19 23 26 19 22 23 HU 17 19 20 17 19 19 MT : 5 : : 4 :

NL 17 18 18 15 17 17 AT 19 22 26 19 22 26 Pl : 8 8 : 7 7

PT 7 8 12 5 7 11

RO 12 13 15 11 12 13 SI : 8 8 : 8 6

SK : 11 12 : 9 11

FI 18 20 : 18 21 :

SE 20 19 18 20 21 22 UK 27 : : 26 : : * FX is Metropolitan France Source: Eurostat demographic data.

As unmarried cohabitation has become a more common living arrangement, the breakdown of cohabitation is also becoming more prevalent. Although reliable comparative data for unmarried partnership dissolution are almost non-existent, the sparse information available from national surveys suggests that unmarried cohabitation is more fragile and of shorter duration than marriage (often less than two years). In the UK for example, data from the British Household Panel show that 60% of consensual unions in the 1990s are known to have turned into a marriage within 10 years, while 35% were dissolved, confirming that unmarried cohabitation was still widely seen as a prelude to marriage. Cohabiting unions that were not converted into marriages were the most likely type of partnership to dissolve. However, marriages with no prior experience of cohabitation were not more likely to breakdown than marriages that began as a cohabiting relationship

In the 1970s, institutionalised life-long marriage contracts were the only legally recognised partnership arrangement in which to give birth and raise children, marriage was most likely to end with the death of one of the partners, and divorce was non-existent or difficult to obtain. In 21

century, couples have a much wider range of options to choose from and are more

likely to experience several different partnership arrangements. These options are not, however, equally available for men and women in all EU-27 Member States. Trend data suggest that, although change is following the same general direction, a large gap remains between the leaders and the laggards.

29 Hantrais L. (2004) Family Policy Matters , The Policy Press, see page 63

2.2.2. Changing patterns of births within and outside marriage

Despite the formal definition used by demographers of a family nucleus as a couple with or without children, for many people, family formation is dependent on the birth of children. As shown in Chapter 1, compared with the 1980s, the transition to parenthood, like that to marital status, has been postponed, and childbearing is more often being compressed into a smaller number of years at a later age, particularly among well-educated women: first births are now more likely to occur as women approach the age of 30, if not later, compared with age 25 in the early 1980s, although some levelling off is occurring.

The extent to which conceiving children is determined voluntarily depends to a large extent on access to effective means of birth control. Information is not routinely and consistently collected on contraceptive use, abortion and childlessness, and little is known about the extent to which childlessness is voluntary or involuntary. Statistics on contraceptive practices do not always include unmarried women, and abortion rates usually record only legal abortions. Where access to a legal abortion is restricted, as in Ireland and Poland, the number of abortions to nationals of the countries concerned may go unrecorded. Data on childlessness are not routinely recorded in many countries and can only be collected at the end of the reproductive life span of a cohort. There is no measure on childlessness that would be equivalent to total period fertility which provides an estimate of the number of children women are likely to have (see the discussion in Chapter 1 on different fertility indicators and postponement).

Although it is not possible to make systematic comparisons of trends over time and within or between countries, the limited information available suggests that more couples in the early 2000s, compared with half a century earlier, are able to control their fertility and choose the number and timing of births using reliable methods of contraception and/or abortion. Voluntary postponement of childbirth is known, however, to increase the risk of permanent childlessness, despite the wider availability of assisted reproductive technology (see Chapter 1).

As more couples form consensual unions and more marriages end in divorce, marriage and parenting are increasingly becoming disconnected. Since cohabitation has replaced marriage as the first form of partnership, many couples are not married when their first child is born. While cohabiting couples often opt for marriage when they decide to have children, a growing number of cohabiting parents choose to raise children outside marriage. As illustrated by Figure 2.2, an important trend in family formation in Europe over the past 40 years has been the large increase in the number of extramarital births. In 2005, about 35% of all European children were born outside a formal marriage, compared with only 8% in 1975.

Figure 2.2: Trends in extramarital birth rates

Live births outside marriage, % of total number of births, average of EU Member States

40%

35%

30%

25%

20%

15%

10%

1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005

Source: Eurostat demographic data.

This phenomenon, which, like unmarried cohabitation, started in the Nordic countries in the 1970s, had spread to all EU Member States by the earlier 21

century. Table 2.7 shows that

Cyprus, Greece, Italy and Spain, but also the Netherlands, were recording very low rates of extramarital births in 1975, whereas the phenomenon was already widespread in Denmark and Sweden. The Southern European countries experienced a large increase in the number of extramarital births over this period, but their average levels remained below those attained in the north of Europe in 2005. The gap between the countries with the highest and lowest rates increased over the period. Cyprus and Greece continued to display rates that were 10 times below the levels in the Nordic countries in 2005. In Sweden and Estonia, more than 55% of births now take place outside marriage (58% in Estonia). Everywhere, the proportion of extramarital births continued to increase between 2000 and 2005. As more countries are officially recognising same-sex couples, some are also allowing same-sex couples to adopt children, and more women in lesbian relationships are bearing their own children

Table 2.7: Trends in the proportion of live births outside marriage, in %

1975 1980 1985 1990 1995 2000 2005 EU-27 : : : : 21.8 : :

BE 3.1 4.1 7.1 11.6 17.3 : :

BG 9.3 10.9 11.7 12.4 25.7 38.4 49.0 CZ 4.5 5.6 7.3 8.6 15.6 21.8 31.7

DK 21.7 33.2 43.0 46.4 46.5 44.6 45.7 DE 8.5 11.9 16.2 15.3 16.1 23.4 29.2 EE : : : 27.2 44.2 54.5 58.5

IE 3.7 5.9 8.5 14.6 22.3 31.5 32.0 GR 1.3 1.5 1.8 2.2 3.0 4.0 5.1

ES 2.0 3.9 8.0 9.6 11.1 17.7 26.6

FX* 8.5 11.4 19.6 30.1 37.6 42.6 47.4 IT 2.5 4.3 5.4 6.5 8.1 9.7 15.4 CY 0.7 0.6 0.4 0.7 1.4 2.3 4.4

LV 11.7 12.5 14.4 16.9 29.9 40.3 44.6 LT 6.2 6.3 7.0 7.0 12.8 22.6 28.4

LU 4.2 6.0 8.7 12.8 13.1 21.9 27.2 HU 5.6 7.1 9.2 13.1 20.7 29.0 35.0 MT 1.2 1.1 1.2 1.8 4.6 10.6 20.0

NL 2.1 4.1 8.3 11.4 15.5 24.9 34.9

AT 13.5 17.8 22.4 23.6 27.4 31.3 36.5 Pl : : : : 9.5 12.1 18.5

PT 7.2 9.2 12.3 14.7 18.7 22.2 30.7 RO : : : : 19.7 25.5 28.5

SI 9.9 13.1 19.1 24.5 29.8 37.1 46.7 SK 5.2 5.7 6.6 7.6 12.6 18.3 26.0

FI 10.1 13.1 16.4 25.2 33.1 39.2 40.4 SE 32.8 39.7 46.4 47.0 53.0 55.3 55.4 UK 9.0 11.5 18.9 27.9 33.5 39.5 42.9 * Metropolitan France Source: Eurostat demographic data.

Figures for extramarital births include births to cohabiting couples as well as those to lone parents. The increase in lone parenting is associated with higher extramarital birth and divorce rates. Even though the great majority of extramarital births take place within cohabiting partnerships, the number of children aged 0 to 14 living with lone mothers has also increased since the 1970s, ranging from 25% in Estonia and 23% in the UK, to 7.9% in Greece and 5.2% in Cyprus according to 2001 census data. This trend can be explained not only by the rise in divorce rates and the breakdown of unmarried cohabitation, but also because of the increasing number of women who decide that they would like to have a child without living with a partner and without jointly registering the birth with the father.

In the mid-1990s, for example, divorce and separation explained more than 50% of lone motherhood, while widowhood explained 20% of lone-parent families across EU-15. In southern Europe, widowhood explained almost 30% of lone parenting, whereas in Denmark more than a third of all lone parents had never been married. Never-married lone parents accounted for 25% or more of lone parents in Austria, Finland, France, Germany, Ireland and the UK.

Like unmarried cohabitation, lone parenthood is not a stable state. In the UK, for example, data from the Office of National Statistics and British Household Survey estimate that in about 50% of cases, lone parenthood lasts no more than four years. It remains, however, that children born to lone parents are likely to spend more years living with a single parent than

children born within an unmarried cohabiting union, who, in turn, spend longer living with one parent than the offspring of a married couple after divorce

For most couples, the birth of children is a reason for getting married. Nevertheless the number of cohabiting couples with children appears to be rising. Table 2.8 presents the importance of various family living arrangements for households with children under the age of 25. In 2001, 80% of all such households with dependent children were headed by a married father and mother; another 6% by cohabiting parents. Single parent families represented around 14% of households with children. Large differences were found across the Member States. The highest levels of single parenthood were found in some of the Eastern European Member States, whereas unmarried cohabiting parenthood was more common in some of the Northern and Western European countries.

Table 2.8: Family status in households with children under the age of 25 (year 2001)

Single Single mother Single parent Married Cohabiting Two adults % of all households with children under 25

fathers with children with children couple couple with children

1 2 3=1+2 4 5 6=4+5

EU-22* 2 12 14 80 6 86 CZ 3 17 20 76 4 80 DK 1 7 8 72 20 92 DE 2 8 10 81 9 90

EE 2 23 25 59 16 75 IE 3 14 17 75 8 83 GR 2 10 12 85 3 88 ES 3 12 15 80 5 85 FR 2 10 12 88 : 88 IT 2 11 13 84 3 87 CY 1 6 7 91 2 93

LV 3 29 32 62 6 68 LT 2 19 21 73 6 79 HU 2 14 16 75 9 84 NL 2 7 9 76 15 91

AT 2 14 16 74 10 84 PL 2 17 19 79 2 81 PT 2 10 12 82 6 88 RO 2 11 13 80 7 87 SL 3 16 19 73 8 81 SK 2 15 17 81 2 83

FI 2 11 13 68 19 87 UK 2 14 16 70 14 84

*EU average calculated for 22 countries, no information is available for BE, SE, MT and LU; no information on unmarried cohabitation for FR. Source: Eurostat 2001 Population Census, own calculations, see also the report for the Commission "Literature review on the impact of family breakdown on children", University of Nottingham.

Compared with the 1970s, more children in the EU are likely to experience transitions between different family living arrangements, to live with only one parent or with stepparents in reconstituted families, and to be raised by mixed-nationality couples (see Chapter 1). However, even in countries with high rates of lone parenthood and divorce, at least two-thirds of children still spend the greater part of their childhood living as a family with both their natural parents. Much of the information presented in this section on unmarried cohabitation and lone parenting is derived from data collected in the 1990s and in the 1991 and 2001

30 Hantrais L. (2004) Family Policy Matters , The Policy Press, see page 68.

censuses. More complete and reliable datasets are needed to track trends in family formation, dissolution and reconstitution with a greater degree of accuracy.

2.3. Changes in labour force participation of women

Arguably, one of the most important trends of the last 40 years affecting family life has been the dramatic increase in women's employment. Since the 1960s, more women have become economically active and have entered paid employment outside the home, particularly in the public services sector, rather than working on the land or in a family enterprise as in the past. Their employment rates have, thereby, moved closer to those of men making them much less dependent on the formal marital relationship for their livelihood. Although the overall trend is for more women with young children to remain in employment during their childrearing years, striking differences can be observed between EU Member States.

Table 2.9 shows the increase in employment rates since 1985 for women aged 25-54, the period in their lives when they are most likely to be combining paid work with family responsibilities. Despite the fact that, for the EU as whole, the employment rate for women aged 25-54 has increased, four groups of countries can be distinguished. In the Northern European countries, employment rates were already high in the 1980s and they have remained above the EU average. In the Western European countries, overall rates were generally lower, but they have since increased markedly, due in some cases, the Netherlands in particular, to the widespread use of part-time work. The third group is formed by the Southern European countries, which reported a relatively low level of women's employment in 1985, and have since seen a marked increase. Although no comparable data are available about women's employment rates in the Central and East European Member States for the years 1985, 1990 and 1995, it is widely accepted that women's employment rates under the socialist regimes were higher than in EU-15 and that, after the transition to a social market economy, women's employment rates fell steeply in many of these countries.

Table 2.9: Employment rates of women aged 25-54, 1985-2005

1985 1990 1995 2000 2005 EU-27 : : : 66.3 69.2

EU-15 : : 61.0 65.7 69.6

BE 48.3 54.5 60.6 67.8 70.4 BG : : : 67.4 70.3 CZ : : : 73.7 74.0

DK 77.0 80.3 75.9 80.4 80.6 DE 54.4 61.5 66.3 71.1 72.5 EE : : : 74.4 77.5

IE 30.3 38.7 48.9 62.5 67.3 GR 43.6 47.1 49.0 52.9 58.5 ES : 37.0 40.2 50.9 61.5

FR 62.2 64.6 67.6 69.6 74.0 IT 42.3 46.2 46.9 50.7 57.9 CY : : : 64.0 72.2 LV : : : 71.7 75.3 LT : : : 76.5 78.8

LU 41.6 48.7 50.6 63.0 68.4 HU : : : 66.7 67.2 MT : : : 32.9 35.4

NL 40.4 51.6 60.5 70.9 75.5 AT : : 69.8 73.5 76.0 PL : : : 64.5 63.1

PT : 63.4 68.9 73.9 74.9 RO : : : 72.7 66.5 SI : : : 79.6 81.1 SK : : : 69.3 69.2

FI : : 71.5 77.6 79.0 SE : : 82.1 80.9 81.1

UK 61.1 68.6 69.5 73.1 74.8 Source: Eurostat, Labour Force Survey.

Table 2.10 looks at trends in the employment gap between women and men. In all the EU Member States for which datasets are complete, the gap has decreased over the period. However, between 2000 and 2005, the gap increased in the countries where it had previously been very small: Estonia, Latvia, Lithuania, Slovenia and Sweden. The marked increase in the employment rates of women in Greece and Spain resulted in a reduction in the gender gap. The reduction over the whole period was faster for the younger than for the older age group, reflecting greater proximity in patterns of male and female employment among younger people and a greater propensity for younger cohorts of women to remain in the labour force.

Table 2.10: Gender Gap in Employment Rates, in percentage points*

Persons aged 25-49 Persons aged 50+

1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 EU : 32 24 21 16 : 24 19 17 15 EU-27 : : : 19 16 : : : 16 15

EU-15 : : 24 21 17 : : 19 17 15

BE 38 32 24 18 14 27 21 18 16 16 BG : : : 5 6 : : : 11 11

CZ : : : 17 18 : : : 19 19

DK 10 7 10 8 8 19 19 18 12 13

DE 34 28 20 16 11 27 27 20 16 12 EE : : : 6 7 : : : 12 9

IE 48 42 30 24 21 35 33 32 29 23 GR 47 43 39 34 29 30 28 25 23 22 ES : 47 37 33 24 : 28 23 23 21

FR 27 25 19 18 14 18 14 11 10 9

IT 49 42 37 33 28 31 28 23 21 19 CY : : : 27 18 : : : 29 28 LV : : : 4 7 : : : 16 14

LT : : : -1 5 : : : 11 14

LU 51 43 39 28 23 27 27 24 21 16 HU : : : 13 15 : : : 13 11 MT ; : : 55 52 : : : 35 33

NL 44 36 26 20 14 27 25 21 21 18 AT : : 19 15 13 : : 20 19 15 PL : : : 14 13 : : : 14 14

PT : 27 19 14 11 : 27 22 20 15 RO : : : 12 13 : : : 12 12 SI : : : 2 5 : : : 18 17

SK : : : 10 13 : : : 16 19 FI : : 5 8 7 : : 7 8 7

SE : : 2 4 6 : : 9 10 10

UK 26 21 16 15 13 22 19 15 14 14 * Male rate minus the female rate Source: Eurostat, Labour Force Survey.

These changes in female labour force participation also have profound consequences for both family and household living arrangements. A significant trend has been the reversal in the correlation between fertility and female labour force participation rates across OECD

. Until

the mid-1980s, the correlation was consistently negative: countries with high average female participation rates displayed lower period fertility rates. During the late 1990s, the correlation started to become positive in some countries. In the early 2000s, the EU Member States with the highest female participation rates also displayed the highest fertility rates. Nevertheless, comparisons of employment rates for women with young children and part-time employment demonstrate that women are still much more likely than men to adapt their employment patterns when they have children.

The combined effect of the trends observed in family living and working arrangements in clusters of EU Members States in the early 2000s can be summarised in the following terms: In Northern Europe, relatively high employment rates of women and small genderemployment gaps can be observed, together with relatively high fertility, indicating that women are more successful in managing to combine family responsibilities and paid work than elsewhere in the EU. The picture is more variegated in Western Europe, where France, the Netherlands and UK display relatively high employment of women and total fertility rates, while above-average employment rates in Austria and Germany are accompanied by low

31 D'Addio A. C. and M. Mira d'Ercole (2004) "Trends and determinants of fertility rates and the role of policies" OECD ( www.oecd.org/dataoecd/7/33/35304751.pdf )

fertility rates. By contrast, the southern European countries, except Portugal and, in some respects, Spain, but including Cyprus and Malta, are characterised by a combination of persistently low levels of female employment, irrespective of whether women have children, relatively large gender employment gaps and low fertility rates, meaning that few women are managing to combine employment with childraising.

Whereas employment rates of women were rising in Western Europe in the 1990s, in the Central and East European countries, both male and female rates fell steeply during the process of economic transition; they were rising again by 2005, albeit with more difficulty in Hungary and Poland. Because female employment rates had fallen from a relatively high level in the 1990s, the gender employment gap continued to be smaller in these countries than in Western Europe. By 2005, as female employment rates began to pick up, fertility rates were also rising.

Although attitudes towards mothers' work outside the home have also been evolving, and Europeans have become more accepting of women combining employment with family life, marked differences remain between countries. Overall, Northern and Western Europe could be said to have achieved a relatively high overall level of labour market integration for women, due often to relatively high part-time rates, combined with widespread acceptance of less conventional living and working arrangements, but nevertheless with substantial variations between countries. Southern Europe and Ireland lie towards the other end of the spectrum in terms of the labour market integration of women and, again, display different approaches towards living and working arrangements. The Eastern European countries are distinguished from most EU-15 Member States by the combination of a relatively small employment gap, low levels of part-time work and, traditionally, a stronger commitment to working mothers in both attitudes and practice, although again with variations between countries.

Marital status, childbearing and childrearing are no longer seen as an insurmountable obstacle for women's employment. Indeed, some countries with high levels of employment of women also have higher fertility rates (see Figure 2.3), suggesting that achieving a satisfactory worklife balance for women may be a key to both higher employment of women and to the maintenance of relatively higher fertility rates, although the strategies adopted in doing so are likely to remain culture specific.

Figure 2.3: Cross-country correlation between employment rates of women and fertility rates

Female employment rate and fertility

1,9

1,8 FI DK UK

1,7 NL t e

r a LU

R 2 = 0,2794 r t i l i t y 1,6 l f e

t a

o EU T

1,5 EE CY AT 1,4

MT PT ES

GR DE IT RO BG LV

1,3 HU CZ LT SK SI PL

1,2

30 35 40 45 50 55 60 65 70 75 Employment rate

Source: Eurostat

2.4. Changing household size and composition

The changes in family life described above, including postponement of marriage and childbirth, lower fertility rates and childlessness, rising divorce rates and the development of alternative modes of family formation, dissolution and reconstitution, are associated with smaller average family size and new patterns of household composition. This section focuses on trends in household living arrangements, with particular reference to household size and composition. The most comprehensive source of data on households is the population census carried out every 10 years in most EU Member States. The last census took place in 2001. Other sources, such as the European Labour Force Survey, and the European Household Budget Survey provide more recent data, but their samples exclude people living in institutions.

2.4.1. Changing household size

As average household size has been declining, the total number of households in Europe has increased much faster over the past 40 years than the size of Europe's population. Several factors have contributed to this development. Firstly, the general decline in the number of births has made families with more than three children increasingly rare. Secondly, the increase in life expectancy combined with the fact that women live on average about six years longer than men means that more women are living alone in old age. Thirdly, the increase in divorce and separation, together with other forms of solo childraising has led to many more single-person households.

Other possible contributing factors include the general increase in economic prosperity, which has made it affordable for people to live in smaller households. Greater prosperity has meant, in particular, that parents and adult children are generally no longer forced to live together under the same roof for economic reasons. Meanwhile, more years spent in education and

later labour market entry for young people have resulted in young adults remaining longer in the parental household.

Table 2.11 combines information from several population censuses to show how household size has changed since 1960

. The table confirms the long-term decline in average household

size, as the number of large households with 5 or more persons has fallen while the number of single-person households has increased. Average household size in 2003 was smallest in Germany and largest in Poland and Slovakia.

Table 2.11: Household characteristics in European countries, 1960-2001

Average household size % of single-person % of households with 5 or households more persons

1960 1980 2001 2003 1960 1980 2001 1960 1980 2001 EU-27 : : 2.5 : : : 29 : : 8

EU-25 3.3 2.8 2.5 2.4 16 21 29 : : 8

EU-15 3.2 2.8 2.4 2.4 16 22 30 21 13 7 EU-10* 3.5 3.1 2.7 : 15 18 26 : : 12 BE 3.0 2.7 : 2.5 17 23 : 16 11 : BG 3.9 3.2 2.7 2.7 6 17 23 : : 9 CZ : : 2.4 2.5 : : 30 : : 5

DK 3.0 2.5 2.2 2.2 20 29 37 15 7 5 DE 2.8 2.4 2.2 2.1 21 31 36 14 8 4 EE : : 2.3 2.6 : : 3 : : 6

IE 4.1 3.8 2.9 : 13 17 22 35 32 18

GR 3.8 3.1 2.8 2.6: 10 15 20 : : 11 ES 3.8 3.4 2.9 2.9 : 10 20 : 26 12

FR 3.1 2.7 2.4 2.4 20 25 31 20 12 8 IT 3.6 3.0 2.6 2.6 11 18 25 27 15 7

CY 3.9 3.5 3.0 3.0 11 10 15 37 25 18 LV : : 2.9 2.6 : : 25 : : 11 LT : : 2.6 2.9 : : 29 : : 8

LU 3.3 2.8 2.5 2.5 12 21 29 19 12 9 HU 3.2 2.9 2.6 2.6 15 20 25 32 17 9 MT 4.2 3.3 : 3.0 11 13 : 37 19 :

NL 3.2 2.5 2.3 2.3 12 22 34 27 12 7 AT 3.0 2.7 2.4 2.4 20 28 34 17 13 8 PL 3.6 3.2 2.8 3.1 16 17 25 : : 14

PT 3.8 3.4 2.8 2.8 11 13 17 29 21 9 RO : : 2.9 2.8 : : 19 : : 14 SI : : 2.8 2.6 : : 22 : : 11 SK : : 3.2 3.1 : : 19 : : 20

FI 3.3 2.6 2.2 2.2 22 27 37 25 10 6 SE 2.8 2.3 : : 20 33 : 13 6 :

UK 3.1 2.7 2.4 2.3 13 22 30 16 11 7 Source: Eurostat Population Census, and Schulz E. (2007) "Household patterns", Research Note prepared for the demography network of the Social Situation Observatory of the European Commission, DIW Berlin Germany.

*

10 Member States that joined in 2004

The 1999 and 2005 Household Budget Surveys provide an indication of more recent changes in average household size, as shown in figure 2.4. In 2005, the average number of persons per household was 2.48 in EU-27, ranging from 2.1 in Denmark and Germany to 3.1 in Cyprus and 3.3 in Malta. Compared with 1999, average household size declined over the period in all EU Member States, with the exception of the UK, where it remained almost constant, and Estonia, Romania and Malta, where it increased.

32 Schulz E. (2007) Household Patterns , Research Note, DIW Berlin, European Observatory on Demography and the Social Situation- Demography Network, European Commission

Figure 2.4: Average household size in European countries, in 1999 and 2005

3,50

3,30

1999 2005 3,10

2,90

2,70

2,50

2,30

2,10

1,90

1,70

1,50

k y d s a a g e l d a s

en lic m ce ia 5) 5) 7) ly ria ia ry nd ia in ia ta ar an an nd tri ni um ur tv ec ga an ki ru al

m ed ub do an on (1 gi (2 (2 Ita ga ga la en tu pa an m nl rla us ua st bo La re ol va S yp M

en er Fi w A ep th ng Fr E on el on on m ul un Ire ov G or P lo om C D G S he Li Ki ni B ni ni B H Sl P S R R xe

et ch d U U U

N te n n n Lu ze ni ea ea ea C U

op op op ur ur ur E E E

Source: Eurostat Household Budget Surveys 1999 and 2005, prepared by DIW Berlin.

Further analysis of the available data by age confirms that young adults and older people generally live in smaller households than people aged 30-59. In EU-27, for the year 2005, average household size of households headed

by people aged 30-44 years was 3.1. For

households headed by people aged 45-59 years, it was 2.7, compared with 2.2 for households headed by young people (aged 20-29) and 1.8 for households headed by older people (60+).

One determinant of average household size is the moment at which young adults decide to leave their parents' home. Information from the 2002 European Labour Force Survey for a limited but representative set of countries, presented in Table 2.12, shows the proportion of young adults living with their parents by gender for four different age groups

33 In most tax systems the head of a family/household is the person in a family or household setting who provides more than half of the financial support to their family/household during the tax year. 34 Fokkema T. and A. Liefbroer (2007) Households in Transition A policy oriented analysis , study cofunded by the European Commission ref. no. VS/2005/0713, NIDI Netherlands

Table 2.12: Proportion of young adults living with their parents in 2002, in %

Women Men

20-24 25-29 30-34 35-39 20-24 25-29 30-34 35-39 BE 65.4 21.4 8.9 5.4 78.5 39.0 17.1 10.8

CZ 69.9 29.5 11.3 7.9 83.7 48.1 22.7 14.6 DE 44.5 13.3 4.8 3.4 62.0 27.2 11.4 7.2

GR 69.7 51.1 25.0 14.4 77.1 71.8 45.9 27.2 ES 82.8 53.5 22.4 13.5 87.5 65.7 32.7 19.6 FR 44.0 10.9 4.6 3.1 62.0 22.6 9.2 6.5

IT 85.0 53.5 23.3 11.3 91.8 72.6 38.6 18.3 LV 64.1 39.4 31.4 16.8 74.7 47.5 34.1 26.4 LT 63.4 38.7 23.7 15.3 74.9 51.7 35.9 19.8 HU 67.9 33.1 17.9 12.8 81.0 51.9 28.7 20.9 NL 43.1 6.3 1.7 0.9 68.3 20.6 5.6 3.3

PT 74.7 45.3 21.3 13.7 83.2 59.9 30.9 19.1 AT 60.1 24.7 10.1 8.7 75.5 43.4 23.9 15.4

SL 85.3 55.2 29.0 17.6 91.2 74.4 46.9 26.9 SK 82.3 55.6 36.3 21.5 89.7 68.3 48.0 34.4 Source: Eurostat 2002 Labour Force Survey, calculations by NIDI

Young adults in Northern and Western European countries are leaving the parental home earlier than in other EU Member States, whereas young people in Southern Europe tend to stay longer with their parents. The same data confirm that in Northern and Western European countries a larger proportion of men and women of all ages tend to live in a single-person household. In Southern Europe, living in a single-person household is not only much less common for young adults, but also less common at older ages. Because of the extended period that parents and children live together, average household size tends to be larger in Southern Europe.

In Central and Eastern Europe, the situation is more heterogeneous. Slovenia, Hungary and the Czech Republic are beginning to look more like the countries in Northern and Western Europe, but the two Baltic States in the sample (Latvia and Lithuania) and Slovakia bear more resemblance to Southern Europe. In sum, Northern and Western Europe, on the one hand, and Southern Europe, on the other, are following a diverging trend, whereas the situation is less clear cut in Central and Eastern Europe.

2.4.2. Changing household composition

Growing numbers of individuals are living alone for various reasons and are not, therefore, classified in census counts as family units, although they may have strong family ties with people living outside their household. Figure 2.5 shows that in 2005 the proportion of singleperson households among all households for EU-27 was 27.7%, ranging from 8.7% in Malta to 40% in Finland. The proportion of family households, defined here as two or more adults living with dependent children, was the smallest in Sweden with 19% and largest in Malta with more than 50%.

Figure 2.5: Proportion of single-person and family households in 2005, in %

s l a e a ly a a y k d

ta in ia ia nd ria ry nd lic 7) 5) m 5) ia ce en al ru ga ki ec ni tv Ita um ni tri an ar an

pa tu an va la ga ga la ub (2 (2 do (1 on gi an ed m M S yp re ve La st el ua us m w nl

C or om lo G lo Po ul un Ire ep on on ng on E Fr th A er en Fi P R S S B H ni ni Ki ni B S D R Li G

ch U U d U an an te an ze ni

C pe pe U pe ro ro ro Eu Eu Eu

Share of single person households Share of family households

Source: Eurostat, Household Budget Survey 2005

A comparison of living arrangements by gender shows that young women leave home and embark on family formation a few years earlier than men. Men tend to lag behind women with regard to life cycle transitions. Another significant gender difference is that very few men live as lone parents. After a divorce, the children usually live with the mother. Due to the longer life expectancy of women, men are also more likely to live with a partner in old age.

People over 80 have a much higher risk of being impaired in their daily living. When this happens, they must rely on personal care and help with housekeeping from other people. Those who still have a partner are less likely to move to a nursing home than widowed or single older people. The 2001 population census contains information about the living arrangements of people by age groups for all EU Member States with the exception of Sweden and Malta. Table 2.13 shows that in EU-27, over 90% of people aged 80-89 and almost 80% of those aged 90+ were still living in private households. This represents around 14.8 million people aged 80+ in the EU-27 still living in a private household. Among them, 46% of those aged 80-89 and 51% of those aged 90+ were living alone.

Table 2.13: Proportion of oldest old (80+) living in private households and institutions in 2001, in %

People aged 80-89 living in People aged 90+ living in

Private Institutions Private Institutions

HH Total Medical Residential HH Total Medical Residential Institutions Homes Institutions Homes

EU-27 92.1 7.9 : : 77.7 22.2 : : EU-15 91.3 8.7 : : 75.6 24.3 : : EU-12 96.9 3.1 : : 94.1 5.9 : :

BE 85.3 14.8 0.7 12.8 61.4 38.5 1.9 34.7 BG 99.1 0.9 0.1 0.8 98.5 1.5 0.2 1.3

CZ 94.0 6.0 0.3 5.3 86.5 13.5 0.4 12.1 DK 93.4 6.6 : : 80.1 19.9 : : DE 90.9 9.1 71.9 28.1

EE 96.1 3.9 0.1 3.5 92.0 8.0 0.4 7.5

IE 84.9 15.1 4.1 8.8 65.2 34.8 7.9 24.3 GR 96.2 3.8 0.8 1.5 94.4 5.6 1.1 2.7 ES 95.1 4.9 0.6 2.8 90.6 9.4 1.2 5.7

FR 87.0 13.0 1.7 10.5 66.9 33.1 4.7 27.6 IT 95.8 4.2 0.1 3.1 89.9 10.1 0.2 8.8

CY 90.4 9.6 0.5 8.8 76.8 23.2 1.4 21.5 LV 97.5 2.5 0.0 2.5 96.2 3.8 0.0 3.8 LT 97.8 2.2 0.1 2.0 96.6 3.4 0.2 3.0

LU 82.6 17.4 1.8 12.0 61.4 38.5 2.9 29.5 HU 94.5 5.5 0.6 4.7 90.5 9.5 1.0 8.1 MT** : : : : : : : :

NL 83.8 16.2 3.3 12.5 55.1 44.9 7.7 36.5 AT 90.1 9.9 1.9 7.6 75.8 24.2 5.2 18.5 PL 97.6 2.4 0.4 1.6 95.4 4.5 0.9 3.1

PT 90.9 9.1 0.3 7.9 82.3 17.7 0.6 15.4 RO 99.3 0.7 0.2 0.4 98.7 1.3 0.4 0.8 SI 91.3 8.7 86.3 13.7

SK* 94.2 5.4 0.7 4.2 91.8 7.7 0.8 6.4

FI* 88.5 7.5 2.1 4.4 68.4 22.2 6.1 13.4 SE** : : : : : : : :

UK 90.4 9.6 4.3 4.4 69.0 31.0 12.9 15.8 * Rows do not always add up to 100% due to unknown arrangements or categories not covered. ** No data for MT and SE Sources: Eurostat 2001 Census data, calculations by DIW

In EU-27, according to 2001 census data, about 1.7 million people aged 80+ and 560,000 people aged 90+ were living in an institution, of whom 83% and 86%, respectively, were women. The proportion of people living in institutions was much higher in EU-15 (9% of those aged 8089 and 24% of those aged 90+) than in the EU-12 (3% and 6% respectively). Moreover, in Northern and Western European countries, people aged 90+ were more likely to be institutionalised than in Southern European countries, where care for older people is mainly provided by families.

A relatively new development particularly in Italy, Spain and Greece has been the arrival of female migrants, notably from Eastern Europe and third countries that often live in the homes of dependent older people and provide the necessary care. According to the Italian National Institute of Social Security (INPS), at the end of 2002, non-nationals represented 56% of the 224,000 registered workers employed in the personal care sector in Italy, and 90% of these non-nationals were women from Eastern Europe and South America.

2.4.3. Projection of future household patterns

The trend towards smaller households can be expected to continue as a result of population ageing. An extrapolation based on the results of the 2001 population census and using Eurostat´s new population projections (EUROPOP2008 convergence scenario, see Chapter 1) is presented in Table 2.14 below. The projection is based on the assumption that household distribution by age group will remain the same as in 2001 and that there will be no other factors (cultural, behavioural, economic) that would lead to different household patterns. Any variations in household patterns would therefore result purely from the changing age structure of the population

The findings show that average household size for EU-27 is likely to decrease further from a level of 2.5 in 2005 to 2.3 in 2050. In 1960, average household size for EU-25 was still at a level of 3.3. The expected demographically induced decrease in average household size between 2001 and 2050 is, therefore, much smaller than the decrease that occurred between 1960 and 2001.

Table 2.14: Projection of household size to 2050 on the basis of EUROPOP2008 convergence scenario

Total HHs % of HH by size according to number of persons Average

in 1 2 3 4 5+ HH size millions

2001

EU-27 187.5 28.8 29.9 17.7 15.5 8.1 2.5 EU-25 177.2 29.3 30.1 17.4 15.3 7.9 2.5 EU-15 150.1 29.9 31.0 17.0 14.9 7.3 2.4 EU-12 37.4 24.6 25.7 20.3 17.8 11.6 2.7 2050

EU-27 219.0 35.0 31.7 14.8 12.1 6.5 2.3 EU-25 209.4 35.2 31.7 14.7 12.0 6.3 2.3 EU-15 181.1 35.2 32.0 14.6 12.0 6.1 2.2 EU-12 37.9 33.8 29.9 15.8 12.2 8.2 2.4 Source: Eurostat 2001 Census data, 2050 calculation by DIW

The number of one- and two-person households is expected to increase considerably. Figure 2.6 shows that in most European countries the number of these small households is expected to increase by at least 50%, and in several countries the number could double between 2001 and 2050.

Schulz E. (2008) Demographic change and the demand for housing , Research Note, DIW Berlin, European Observatory on Demography and the Social Situation- Demography Network, European Commission

Figure 2.6: Increase in the number of one- and two-person households between 2001 and 2050

200

1-p-hh 2-p-hh 150

100

y a k d 2 e l 0 d a s a g s

ce ia ia ia ry 15 m ly 27 25 ia ria lic in nd

ar tv an ni ar an um S1 ec ga S1 an ni nd ki ur ru

an on m ga U do Ita U U gi an st tu ve ub pa va la lg La st m ua E E E nl M re M ol rla S bo yp

Fr E er th en un ng Fi el om N G Au or N P lo ep lo m C Ire Bu H Ki B R P S he S G Li D R xe d et

te ch N Lu ni ze U C

Source: 2001 Eurostat Population Census, 2050 DIW calculation using Eurostat, EUROPOP2008 convergence scenario.

The increase in the number of single-person households will be particularly large due to the rising number of older people living alone: from 32.3 million in 2001 to around 51.6 million in 2050 (see Table 2.15). The number of people aged 80+ living alone will increase most rapidly: from 6.1 to 22.5 million.

Table 2.15: Number of older people, total and singles, in 2001 and 2050, in millions

2001 2050

Total Singles Total Singles 60 +

EU-27 111.5 32.3 170.9 51.6 EU-15 90.2 26.8 135.8 42.0 EU-12 21.2 5.5 35.0 9.6 80+

EU-27 13.0 6.1 48.8 22.5 EU-15 10.8 5.2 39.9 19.1 EU-12 2.2 0.8 8.8 3.4 Source: 2001 Eurostat Population Census, 2050 DIW calculation using EUROPOP2008 convergence scenario.

2.5. Adapting policies to changing family and household patterns

Governments use data on changing social realities to inform public policies and to formulate policy measures designed to ensure that the needs of the population are still adequately met. This implies that they keep under review rights and obligations under family law, the design

of tax-benefit system, and the provision of housing and social services, especially child and elder care, which are crucial in supporting women's labour force participation.

How successful policy adaptation is can be monitored using a range of indicators, such as those that have been developed for the Open Method of Coordination (OMC) on social protection and social inclusion. Many of these focus on poverty risks of various groups within the population, defined by gender, age, household type and labour market involvement, in particular

. The OMC has now been in operation for 8 years and has resulted in a much better

understanding of social challenges, due to the availability of this battery of agreed common indicators.

These indicators show that much more needs to be done, in many Member States, to protect one of the family types that could become more prevalent as a result of the greater variety of family forms described in this chapter, in particular single-parent families. These consist mostly of women with their dependent children. Around one-third of these families are at risk of poverty (see Figure 2.7), twice the proportion for the population as a whole. The problem is being addressed through higher benefits, and through measures that enable single parents to reconcile regular employment with family responsibilities.

Figure 2.7: Exposure of single-parent families to the risk of poverty, 2006

At-Risk-of-Poverty Rates: Single-Parent Families

Total population

Single parents

% 25 20

Z K E Y U T L T T O I K

25 BE BG C D D EE IE EL ES FR IT C LV LT LU H M N A PL P R S SK FI SE U U

Source: SILC 2006, Income data 2005; except for UK, income year 2006 and for IE moving income reference period (2005-2006); BG and RO: National HBS 2006 and income data 2006, PT provisional values.

However, it is not only lone-parent families that are exposed to a high risk of poverty. Couples with three or more children a family type that is becoming less frequent are also more vulnerable than the population in general: 24% of these large families have an income

36 See also the 2008 Joint Report on Social Protection and Social Inclusion.

below the poverty threshold , compared to 16% for the population as a whole. Some

countries do have policies in place that protect large families as well as the general population: in Belgium, Denmark, Germany, Cyprus, Finland and Sweden, couples with three or more children experience poverty hardly more often or even less often than the general population. In most Southern, Central and East European Member States, by contrast, large families are twice as much at risk of poverty as the population as a whole (see Figure 2.8).

Figure 2.8: Exposure of large families to the risk of poverty, 2006

At-Risk-Of-Poverty Rates: Large Families

Total population

Two-adult households with 3+ children

E Z K E E IE L S IT Y U T L L O S I K FI E K

25 B BG C D D E E E FR C LV LT LU H M N AT P PT R S S U U

Source: EU-SILC 2006 (income data for 2005); BG and RO: National HBS 2006 and income data 2006

About 19% of children up to the age of 18 were at risk of poverty in the EU in 2006, meaning that the disposable income of the household in which they live, adjusted for household size, was below 60% of the national median income (compared to an at-risk-of-poverty rate for the whole population of 16%). All families with children are therefore at a higher risk of poverty than the population in general. Figure 2.9 shows that the risk of poverty is highest in several Southern and Central and East European countries: a quarter or more of children are concerned in Poland, Latvia, Italy, Lithuania and Hungary, and just under a quarter in Spain, Italy, Lithuania, and the UK. Two Nordic countries, Finland and Denmark, have the lowest risk of poverty at 10%.

People are regarded as being at risk of poverty if their income, adjusted for household size, falls below 60% of the national median income.

For details see http://ec.europa.eu/employment_social/spsi/common_indicators_en.htm

Figure 2.9: At-risk-of-poverty rate after social transfers as % of children, in 2005

At-risk-of-poverty after transfers of children, in 2006

EU average

L I Y K IT LT HU ES UK EL IE PT EE LU MT SK BG CZ BE AT SE FR NL DE S FI LV P C D

Source: Eurostat, EU SILC 2006 (income data for 2005).

The financial situation of families and the risk of poverty they are exposed to depend on the combination of income that parents can earn and the benefits they receive. Benefits may go a long way towards covering the costs of raising children, but would have to be very high to replace a second income in a two-adult household. The availability of affordable childcare, enabling both parents to participate in the labour force, can therefore have a major impact on the financial situation of families.

Box 2.2: Public opinion on priority actions for families

A recent Flash Eurobarometer 247 asked EU citizens about the most important policy

measures that could improve life for families. More flexible childcare arrangements and increased tax advantages received the strongest support. The demand for flexible childcare was particularly high in some southern European Member States.

Question: Various policy measures can help improve life for families. For each of the measures I am going to read out, would you say it should receive high, medium or low priority for policy action in [COU TRY]?

Priority actions to improve the life of families

Acce ss to more fle xible childcare arrange me nts 66 24 5 5

Incre ase d tax adv antage s for familie s with childre n 66 22 8 4 Imple me nting the right of paid le av e to care for

60 28 8 4 de pe nde nt e lde rly pare nts and re lativ e s

Incre ase d child allowance s 59 27 11 4

Easie r acce ss to part-time work 58 30 9 4 Longe r paid pare ntal le av e 47 32 15 5

Ince ntiv e s for fathe rs to take pare ntal le av e 40 35 19 5

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% High priority Middle priority Low priority DK/NA Access to more flexible childcare arrangements

100% 2 3 3 3 4 2 4 4 2 2 3 4 2 6 7 5 6 5 6 7

4 3 4 5 10 8 5 5 6 5 5 9 12 10 10 9 11 11 10 3 4 6 5 5 11 6 90% 16 3 10 4

8 3 7 6 5 6 16 17 21 14

80% 22 23 23 24 28 30

21 22 22 22 24 27 30 70% 23 27 29 26 33 33 34 31

38 41 25

60%

50%

40% 80

77 77

72 69 68

67 67 67 67 67 67 66 65 65 30% 64 63 60

59 57 57 56 53 53 51 50 46 43 20%

10%

T S R Y E L G IE T T IT U 27 R K I E K O F I E E T L K Z P E G C D P B M A H U F U LU S B D LV R S E L N S C E

High priority Middle priority Low priority DK/NA Source: Flash Eurobarometer 247

Some simple correlations seem to confirm the view held by the Commission that reconciliation measures are of crucial importance. Although higher cash benefits per child (in relation to GDP per capita) are positively correlated with large family size, the link is extremely weak (see Figure 2.10). The strongest correlation would seem to exist between the availability of childcare, and higher level of employment of women, which bring more income into families thereby reducing the risk of poverty. Indeed, only about 7% of

The fieldwork was carried out between 10th and 14th September 2008. Over 27,000 randomly selected citizens aged 15 years and above were interviewed in the 27 EU Member States. Interviews were predominantly carried out via fixed telephone.

households with dependent children were at risk of poverty in 2005 when each adult household member was in employment.

A stronger focus on reconciliation measures in supporting families has other advantages: it increases the labour supply, keeping ever better educated women in employment. This increases the growth potential of the economy and strengthens the ability of a country to confront the challenges of an ageing society.

Figure 2.10: Correlations of fertility with indicators related to family and reconciliation policies

Cash benefits and fertility

1,9

1,8 DK FI UK SE BE

1,7 NL t e

r a LU

r t i l i t y 1,6 l f e 2

t a R = 0,0553 o EU T EE 1,5

CY AT

1,4 PT MT

ES DE

IT GR RO

1,3 LV HU LT BG CZ SI

PL SK 1,2

0,0% 2,0% 4,0% 6,0% 8,0% 10,0% 12,0% 14,0% Cash benefits per child as % of GDP/head

Child poverty and fertility

1,9

1,8 FI SE UK BE

1,7 NL t e

r a LU

r t i l i t y 1,6 l f e

t a EU o

1,5 EE CY AT

1,4 PT MT 2

ES R = 0,1743 DE GR

BG HU LV IT 1,3

CZ LT SI SK PL

1,2

5 10 15 20 25 30 % of children at risk of poverty

Availability of childcare for children aged 0-2 and fertility

FR 2

R = 0,4004 1,9

1,8 FI DK UK SE BE

1,7 NL t e

r a LU

r t i l i t y 1,6 l f e

t a EU o

1,5 EE CY

1,4 AT PT MT

DE ES GR IT

1,3 HU LV CZ LT SK SI PL

1,2

0 10 20 30 40 50 60 70 80 Childcare availability 0-2 years

The rising number of single-person households also warrants some attention to the social conditions of people with this living arrangement, particularly in the case of older women. Figure 2.11 shows that, for the EU as a whole, both single men (at 22%) and women (at 25%) are at a higher risk of poverty than the population as a whole (16%). Again, the picture is very heterogeneous across the EU. In some countries, single women or single men (but never both) may be less at risk of poverty than the general population. There are also several countries where single men are exposed to a higher risk of poverty than single women. Each country thus faces its specific policy challenges with regard to the various family and household types.

Figure 2.11: At-risk-of-poverty rate for single women and men, in 2005

At-Risk-of-Poverty Rates: Single Women and Men

Total population

55 Women living alone Men living alone

% 30 25

Z K E IT Y U T L T T O I FI K

25 BE BG C D D EE IE EL ES FR C LV LT LU H M N A PL P R S SK SE U U

Source: Eurostat, EU SILC 2006 (income data for 2005).

Box 2.3: The EU's contribution to better policies for families

· In the context of the Open Method of Coordination on social protection and social

inclusion, the analysis of common indicators contributes to a better knowledge of how well different types of family and household are protected against the risk of poverty. Ongoing and planned work in the framework of the OMC, notably on child poverty and older women, will help build up the knowledge base on the problems they create and the policies that can be used to tackle them.

· The implementation of the Roadmap for equality between women and men adopted by the

Commission in March 2003 (COM(2006) 92), and in particular of the measures under the heading reconciliation of work, private and family life, will also contribute to better conditions for families.

· A European Alliance for Families was established under the German presidency in 2007,

following the adoption by the Commission of the communication on Europe's Demographic Future: From Challenge to Opportunity (COM(2006)571). The Alliance provides a platform for mutual learning for the Member States to help them modernise their family policies and respond to new challenges arising from the social changes presented in this chapter.

· One of the initiatives taken under the Alliance is to cooperate with the OECD on the 39

development of its Family Database and to ensure that this database becomes a tool that

allows all EU Member States to assess the situation of families in their country from a comparative perspective. This work is in progress and will be presented in the 2010 Demography Report.

· Also linked to the European Alliance for Families is the establishment of a Web Portal which will present information on family policies in the Member States and examples of good practice at the local and company level

· The European Structural funds can, under certain conditions, support initiatives in the

Member States with the aim of creating better conditions for families, notably through measures that facilitate the reconciliation of work and private life. A brochure has been prepared in cooperation with the expert group on demographic issues established in June 2007 by the Commission. The funding of a Social Platform on family issues is foreseen under the 7 th · Framework

Programme for research and development. This platform would bring together researchers, policy makers and stakeholders to help orient future research according to the needs discussed with policy makers and stakeholders.

· The Commission assesses the progress towards the targets for childcare set in Barcelona in 41

2002 in a forthcoming Communication

See: http://www.oecd.org/els/social/family/database 40 http://ec.europa.eu/employment_social/families/index_en.html 41 To be completed after adoption.

3.

O PPORTUNITIES AND N EEDS IN AN A GEING SOCIETY

The population pyramids presented in Chapter 1 clearly show the marked increase in cohort sizes just after the end of World War II. Sixty years later, the first of these large cohorts born over a period of 20-30 years are beginning to retire, signalling a turning point in the EU's demographic development: population ageing no longer lies in a distant future. The Commission's Renewed Social Agenda42 identified Europe's ageing society as a priority area, stressing the need for a variety of policy responses. This chapter highlights the potential that the still healthy and fit baby-boom cohorts represent for Europe and explores, as announced in the Renewed Social Agenda, the actions required to meet the needs of an ageing population.

Over the past decade, both the population of working age (20-59 years) and the population aged 60 years and above grew by 1 to 1.5 million people per year on average. From now on, the population aged 60 years and above will be growing at the rate of 2 million people every year for the next 25 years. The growth of the working age population is slowing down rapidly and will stop altogether in about 6 years; from then on, this segment of the population will be shrinking at the rate of 1 to 1.5 million people each year, as illustrated by Figure 3.1.

Figure 3.1: Population change over previous year, EU 27, 1995-2049

3.000.000

2.500.000

2.000.000

1.500.000

1.000.000

20-59 60+ 60-69

500.000

95 97 99 01 03 05 07 09 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 19 19 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 -500.000

-1.000.000

-1.500.000

-2.000.000

Source: calculation based on Eurostat demographic data, including EUROPOP2008 convergence scenario projection from 2009 onwards.

Societies have to adapt to this rapidly changing age structure. This implies first of all offering the increasing number of older people better opportunities to make an active contribution to the economy and to society. In 2007, by the age of sixty, only about 48% of men and 31% of women were still in employment. Yet, most people in this age group are still fit and capable

42 COM(2008) 412 of 2 July 2008.

of contributing to the economy and society. Employment rates of people aged 55-64 are already rising, reversing the past trend towards ever earlier retirement. This is a clear indication that the Lisbon strategy is working even if more needs to be done. Moreover, a recent Flash Eurobarometer opinion poll revealed that three-quarters of respondents would consider participating in community work or volunteering after they retire (see box 3.1 below). It will also be necessary to ensure that older people have adequate incomes in retirement as well as access to goods and services that allow them to preserve, for as long as possible, their autonomy. There is a need to strengthen solidarity with frail older people who have become dependent on the help of others and who, as a result, are particularly vulnerable.

This chapter provides data related to the first of these three policy priorities, i.e. opportunities for an active contribution to the economy and society. New data sources will become available over the coming years that will allow an analysis of access to goods and services and the need for long-term care. Further work is also required on the quality of care and the prevalence of elder abuse and neglect.

3.1. Older people in employment

Between the age of 50 and 70, labour force participation rates decline steadily. For the EU as a whole, 85% of men aged 50 are in employment and 70% of women. By the age of 69 for men and 66 for women, the employment rate falls below 10% (see Figure 3.2). However, a significant change can be observed since the year 2000: employment rates have risen markedly at almost all ages, in particular between 54 and 61 years for women and the early sixties for men. An additional ten percentage points of women and men aged 60 are now in employment, compared to 2000. Due to the baby-boom, these cohorts are particularly large, so that an increase in the employment rate will have a strong impact on total employment.

Figure 3.2: Employment rate of persons aged 50-69 in the EU-27, by gender and age, 2000 and 2007

100,0

90,0

80,0

70,0

t e 60,0 t r a

n Men 2000 e 50,0

m Men 2007 y

l o

p 40,0 Women 2000 m

E Women 2007 30,0

20,0

10,0

0,0

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 Age

43 See also chapter 2 of the 2007 Employment in Europe Report. http://ec.europa.eu/employment_social/employment_analysis/employ_2007_en.htm

Source: Eurostat, Labour Force Survey.

The EU has thus made significant progress towards the target set by the Stockholm European Council within the framework of the Lisbon Strategy, namely to raise the employment rate of people aged 55-64 to 50% by the year 2010. In 2007, 12 Member States had reached the target, among them all the Nordic and Baltic countries (see Figure 3.3). The EU-27 average employment rate for people aged 55-64 increased by nearly eight percentage points from 36.9% in 2000 to 44.7% in 2007. For the prime working age group (25-54), the increase was only 3.1 percentage points, mostly due to the increased labour force participation of women in this age group.

Significant differences exist across Member States as far as the employment of older workers is concerned. Sweden has by far the highest employment rate at 70%, 10 points higher than the countries in second and third position (Estonia and Denmark). Poland and Malta have the lowest rates, below 30%. A few countries are distinguished by the small gap in the employment rate between women and men: Estonia, Finland, France and Sweden.

Figure 3.3: Employment rates of older workers aged 55-64, by gender in EU-27, in %

Employment rates of older workers aged '55-64' by sex and country, 2007

) 60 ( %

t e 50 Total

t r a

e 40 Men m

y Women l o 30 p

SE EE DK LV UK CY FI IE LT DE NL PT CZ EU ES BG GR RO AT FR SK BE IT SI HU LU PL MT Member State

Source: Eurostat, Labour Force Survey.

Figure 3.4 shows how employment rates decline with age. The chart displays employment rates for five-year age classes from 50-54 to 70-74 and illustrates in which of age groups the largest number of labour market exit is most frequent. A significant proportion of both women and men in EU-27 already leave the labour market in their mid-fifties: between the age groups for 50-54 and 55-59, the employment rate falls by 15.7 points in the case of men and 18.3 points in the case of women. The biggest falls for both women and men occur between the age groups for 55-59 and 60-64: 29.3 points for men and 26.9 points for women. Very few stay on the labour market beyond the age of 65, which is the statutory retirement age in many Member States: in the age group 65-69, the employment rate falls by further 25 points for men and 14.7 points for women.

The patterns of labour market exit differ significantly from one country to another. Some countries display large falls in employment rates at a young age: Czech Republic for women, but not for men, Luxembourg for men, but less so for women, Poland, Slovenia and Slovakia for women. Sweden's success in achieving the highest employment rate of older workers is

largely due to the fact that labour force participation remains very high up to the age class 6064 and only drops sharply thereafter. Denmark, which also has a high labour force participation rate for older workers (55-64), by contrast, sees its employment rates already fall sharply between the age groups 55-59 and 60-64.

A handful of countries record employment rates above 20% in the age group 65-69 years. This is the case for women in Romania and Portugal, with Latvia almost reaching 20%. Around 30% of men aged 65-69 are still in employment in Romania, Portugal, Cyprus and Latvia, followed by Estonia and Ireland with about 27%.

Figure 3.4: Employment rates of older workers in EU-27, by age group*, in 2007

Employment rates by age class

Women, 2007

y50_54

y55_59

y60_64

y65_69

y70_74

eu be bg cz dk de ee ie gr es fr it cy lv lt lu hu mt nl at pl pt ro si sk fi se uk

Employment rates by age class

Men, 2007

100

y50_54

y55_59

50 y60_64 y65_69

y70_74

eu be bg cz dk de ee ie gr es fr it cy lv lt lu hu mt nl at pl pt ro si sk fi se uk

Source: Eurostat, Labour Force Survey. For some countries, employment rates at higher ages are not available (the column does not display the corresponding segments in those cases).

*

For some countries, employment rates are not available. In such cases, the bar does not show the corresponding segment. Some employment rates in higher age groups are uncertain or unreliable.

The employment rate data presented above do not take account of the number of hours worked. Figure 3.5 presents the distribution of older male and female workers by the number of hours worked. The majority of men aged 55-64 worked 40 hours or more per week, whereas, in most Member States, only a minority of women worked so many hours. In 2007, 22.2% of all European workers aged 55-64 said that they worked part-time

. For women

aged 55-64, the prevalence of part-time working was much higher than for men: 38.3% compared to 10.9% for men.

The proportion of part-time employment among older workers is higher than among primeage workers: in the 25-49 age group, 15.7% were working part-time (4.7% of men and 29.1% of women). However, particularly for older men below the statutory retirement age, the choice is typically between full-time work and complete labour market exit. Gradual retirement in the form of part-time working is not yet very widespread.

The vast majority of older part-time workers would not want to work full time. Only 15.5% of older part-time workers said that they had accepted a part-time job because no full-time job was available (the proportion was slightly higher for men than for women: 16.8% compared to 15%).

The classification as part-time worker in the Labour Force Survey is based on spontaneous answers given by respondents as there is no common definition of the working time that would constitute fulltime employment.

Figure 3.5: Working hours (usual) of older workers, aged 55-59, and 60-64, by range of hours and gender, in 2007, EU-27

Workers aged 55-59 Workers aged 60-64 Percentage

Percentage

0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% T M T M A A F F

E M E M B B F F

G M G M B B F F

Y M Y M C C F F

Z M Z M C C F F

E M E M D D F F

K M K M D D F F

E M E M E E F F

S M S M E E F F I I

F M F M F F

R M R M F F F F

R M R M G G F F

U M U M

t e H t e H F F

t a t a

r S I E M r S I E M e e

b F b F m m e e

M I T M M I T M F F

T M T M L L F F

U M U M L L F F

V M V M L L F F

T M T M M M F F

L M L M N N F F

L M L M P P F F

T M T M P P F F

O M O M R R F F

E M E M S S F F

I M I M S S F F

K M K M S S F F

K M K M U U F F

+40 30-39 15-29 <15 +40 30-39 15-29 <15 Source: Eurostat, Labour Force Survey.

A different picture emerges with regard to part-time employment for the few people who remain in employment at the age of 65 and over. In EU-27, more than half of workers aged 65+ work part-time (see Figure 3.6). The figures are particularly high in the Netherlands (where part-time work is common across the age groups, among both women and men), the UK, Sweden, Germany, Finland and Austria.

The pattern of part-time working at the age of 65+ differs much less between women and men. However, the prevalence of part-time working is still lower for men than for women. These data suggest that part-time working can be a useful option for continuing labour force participation.

Figure 3.6: Part-time working among older and prime-age workers, men and women, 2007, in %

70 65+ 25-59

% 50 40

L E I T T T L E I Z Y K E T S V

K E F A K S R U IE L O L IT G R U

N U S D D M P P B 27 C F H C S E E R B G L U

Share of part-time working,

Women, 2007

100

65+

25-59

60 %

L K E E K T T Z L E I K U E Y S O G T R A FI IE P P 27 S LT LV IT

N U D S D C LU B S FR H E C E R B M G U

Share of part-time working,

Men, 2007

65+

25-59

L T I T T I L Z

E F K E K E S R Y U IE S K O IT V E R G U

N M S U A D D B P P F 27 C C H E LT S L E L U R G B E

Source: Eurostat, Labour Force Survey.

In the 50+ age groups, the employment rate depends very much on their level of education (see Figure 3.7). In this age group, the gap in the employment rate between people with the highest level of education and those with the lowest level is about 25 percentage points at the age of 50, and more than 30 points in the late fifties. At the age of 65 or older, people with the highest level of education are twice as likely to be in employment as those with lower levels of education, but only about one in six of those with a high level of education will still be in employment after the age of 67.

Figure 3.7: Employment rates of older people (50-69) in EU-27, by level of education* and age, in 2007

Employment rate of older people (50-69) in the EU27, by level of education and exact

age, 2007

100,0

90,0

80,0

t e 70,0

t r a 60,0 n ISCED 0-2 e

m 50,0 ISCED 3-4

l o 40,0 ISCED 5-6 p

E 30,0 20,0

10,0

0,0

50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 Age

Source: Eurostat, Labour Force Survey.

*

Level of education is coded according to the International Standard Classification of Education 1997. Lower secondary corresponds to ISCED1, 2 and 3c short programmes, upper secondary to ISCED 3a, 3b, 3c long and ISCED 4 programmes, tertiary to ISCED 5 and 6 programmes.

The baby-boom cohorts started to benefit from the expansion of higher education in the 1960s and 1970s. The proportion of people with high levels of educational attainment is higher among younger than among older cohorts. This can be seen in Figure 3.8, which presents the distribution of the population aged 25-59 by level of education. Whereas almost 40% of those aged 55-59 had a low level of education, only about 20% of those aged 25-34 were in this situation. The reverse can be observed for tertiary-level education: 20% of those aged 50-59 had a university education, compared to 30% of those aged 25-34. The proportion with intermediate levels of education also increases in the younger age cohorts.

The trend towards higher levels of educational attainment is particularly pronounced in the case of women. The proportion of women with a low level of education is high in the oldest age class: 43.2% compared to 33.5% of men. The situation is very different for younger women aged 25-29: only 17.6% have the lowest level of education, compared to 21.5% of men. The same reverse relationship can be observed for the highest level of education: 34% of women aged 25-29 have completed tertiary education, compared to 25.3% of men in the same age group. In the age group 55-59, by contrast, men have the edge over women with 20.7% of men having achieved tertiary education compared to 16.2% of women.

Figure 3.8: Educational attainment* by age group, women and men, EU-27, 2007

Educational attainment by age group

EU27 women and men, 2007

100,0%

90,0%

80,0%

70,0%

u 60,0% r o

g isced5_6 e

a 50,0% isced3_4 e

isced0_2

f t h

o 40,0% %

30,0%

20,0%

10,0%

0,0%

y25_29 y30_34 y35_39 y40_44 y45_49 y50_54 y55_59

Educational attainment by age group

Educational attainment by age group EU27 men, 2007 EU27 women, 2007

100,0%

90,0%

80,0%

70,0%

60,0% 60,0% 50,0% 50,0% 40,0% 40,0% 30,0% 30,0% 20,0% 20,0% 10,0% 10,0% 0,0% 0,0%

y25_29 y30_34 y35_39 y40_44 y45_49 y50_54 y55_59 y25_29 y30_34 y35_39 y40_44 y45_49 y50_54 y55_59 Source: Eurostat, Labour Force Survey.

*

Level of education is coded according to the International Standard Classification of Education 1997. ISCED: 1 Primary education; ISCED 2: Lower secondary education; ISCED 3: Upper secondary education; ISCED 4: Post secondary education, non-tertiary; ISCED 5: Tertiary education; ISCED 6: Postgraduate education. Lower secondary corresponds to ISCED 1,2,3c short, upper secondary to ISCED 3a, 3b, 3c long , 4 and tertiary to ISCED 5 and 6.

Low digital literacy may increase the likelihood of older workers dropping out of the job market prematurely. Internet usage among senior citizens, measured by the figures for those who have used the Internet in the year preceding the survey, still varies greatly between Member States. The largest take-up across all age groups can be observed in the Nordic countries, as well as in the Netherlands and Luxembourg. Most new Member States and the Southern countries lag behind, some well below the EU-27 average (see Table 3.1).

Internet usage declines sharply from the youngest to the oldest age groups. In the age group 55-64, it is only about half the level recorded for people up to the age of 35, putting many older workers at a disadvantage on a labour market where skills in using computers and the Internet are becoming more and more important. The age gradient in Internet usage is likely to reflect how early in life people are exposed to ICT and the Internet. New cohorts entering

the higher age brackets can be expected to be much more willing to embrace these new technologies and the age gradient can be expected to flatten over time. This can already be seen between 2004 and 2007: the increase in Internet usage has been most marked in the older segments of the population, albeit starting from a very low level. Internet usage rose by 54% among people aged 55-64 and 80% among those aged 65-74 (compared to an overall average increase of 28%) (see Figure 3.9).

Figure 3.9: Internet usage by age group in EU-27 over the past year, 2004, 2007, in %

Internet usage in the last year

(EU27, % per age class, 2004, 2007)

100

2004

2007

Total 16-24 25-34 35-44 45-54 55-64 65-74 population

Source: Eurostat, Information society statistics.

Table 3.1: Percentage of individuals in EU-27 who used the Internet, by age, in 2007

2007 16-24 25-34 35-44 45-54 55-64 65-74 Average EU 27 88 77 69 58 40 18 60 BE 94 85 82 69 49 21 69 BG 67 48 40 27 12 2 34 CZ 86 64 61 50 27 8 52

DK 100 97 94 90 77 46 85 DE : : 90 78 60 : 75 EE 96 90 80 : : : 66

IE 78 78 67 54 32 17 61 GR 77 59 41 23 10 2 36 ES 90 77 61 49 23 7 55 FR : 88 75 66 : : 66

IT 71 59 48 39 21 5 41 CY 70 59 44 30 15 4 41 LV 96 81 69 50 29 7 59 LT 91 69 55 39 17 4 50

LU 98 88 87 81 66 29 79 HU 86 73 61 48 31 8 53 MT 81 67 54 37 21 : 47

NL 99 97 95 91 71 44 86 AT 91 86 79 71 48 23 69 PL 90 69 54 36 20 : 49 PT 88 63 44 : : : 42

RO 60 38 29 20 7 1 28

SI 94 84 68 46 23 7 57 SK 93 76 70 63 22 4 62

FI 100 99 94 87 65 30 81 SE 93 93 90 84 75 44 82 UK : : : : : : 75 `:' indicates unreliable data.

Source: Eurostat, Information Society statistics.

The increasing familiarity with the Internet is also supported by figures showing how many people have never used it. While 80% of those aged 65-74 across the EU have never used the Internet, this figure drops to 57% among the next younger age group, and then again to 39% among those aged 35-44 (see Table 3.2). This signals a marked trend towards more competence in this field of ICT, and hence more and more ICT-capable older cohorts.

Table 3.2 - Percentage of individuals in EU-27 who have never used the Internet, by age, in 2007, in %

2007 16-24 25-34 35-44 45-54 55-64 65-74 Average EU 27 9 20 28 39 57 80 37 BE 5 13 16 29 48 76 29

BG 31 49 58 72 87 98 65 CZ 11 31 38 49 70 91 46 DK 0 2 4 7 20 47 12

DE : : 9 20 38 71 23

EE 3 8 20 39 62 81 32

IE 16 17 28 44 66 81 35 GR 20 37 57 75 89 97 62 ES 8 20 37 49 76 92 43 FR 4 11 23 33 53 81 32

IT 23 36 47 56 74 90 54 CY 25 36 53 69 84 95 56 LV 3 16 30 48 69 91 39 LT 7 30 44 60 82 96 49 LU 2 10 13 18 33 69 20

HU 13 26 38 51 68 91 46 MT 14 30 46 62 78 89 51 NL 0 2 4 8 26 52 13

AT 7 11 18 27 49 74 28 PL 6 26 43 61 77 93 48

PT 10 34 54 71 80 94 56 RO 36 59 69 78 92 99 69 SI 2 13 26 50 72 92 39 SK 5 19 27 34 75 92 35 FI 0 1 4 10 31 65 17 SE 6 5 9 13 20 46 15

UK : 11 16 19 36 60 22 Source: Eurostat, Information Society statistics.

The shift towards higher levels of education and ICT skills means that future cohorts will be more likely to stay on the labour market up to the statutory retirement age and possibly beyond. Particularly older women can be expected to be in a much better position to remain longer in paid work.

While future cohorts of older workers will find themselves better equipped for longer working lives, due to their higher level of educational attainment and, consequently, their ability to keep their skills up-to-date, they may have other reasons for leaving the labour market early. Figure 3.10 presents the main reasons why people aged 55-64 are inactive. The reasons given may, to a large extent, reflect national specificities, and the results should not be regarded as comparable across borders. However, some interesting conclusions can be drawn from the data.

Most people in this age group state that they are retired, implying that they would not envisage a return to the labour market, and that they benefit from social protection arrangements that enable them to retire from the labour market. In many countries, illness or disability constitutes the second most important reason for inactivity. Information coming from register data in the Member States suggests that the rate of chronic illness and disability rises progressively from 1% among young people to 15% at retirement age.

See page 25 of "Compilation of disability statistics from the administrative registers in the Member States", by APPLICA,CESEP &European Centre, November 2007, Study for the European Commission, VC/2006/0229, http://ec.europa.eu/employment_social/index/comp_disb_final_en.pdf

Disability is most important in two countries with high employment rates of older workers, namely Sweden and Finland. This does not imply that people are more likely to be ill or disabled in these countries. A smaller proportion of people in this age group are outside the labour market than in most other countries. Early retirement is not common, and the main reason for leaving the labour market would have to be a health problem, whereas in other countries other labour market exit pathways may be available. Few people in this age group say that they are inactive because they think that no jobs are available.

The third most important motivation for being inactive are reasons related to the family care obligations regarding children or disabled adults, or personal reasons. Here, important differences are found between men and women: only 1.4% of men indicate that they are inactive for family or personal reasons, compared to 16% of women. This may reflect the disproportionate share of care obligations borne by women. It may also reflect the persistence of the male breadwinner model among the older cohorts: if the husband's income is sufficient, a second income is not necessary, and the wife can stay out of the labour market for family or personal reasons.

Figure 3.10: Reasons for inactivity, Women and Men (55-64), 2007

Reasons for inactivity

Women and Men 55-64, 2007

100

60 Other reasons No jobs

50 Family/care/personal reasons Illness/disability

40 Retired 30

27 cz gr fr it cy lv lt lu t nl at pl pt ro si sk fi be bg dk de ee es hu m se eu Reasons for inactivity

Women 55-64, 2007

100

60 Other reasons No jobs

50 Family/care/personal reasons Illness/disability

40 Retired 30

27 cz gr fr it cy lv lt lu t nl at pl pt ro si fi be bg dk de ee es hu m sk se eu Reasons for inactivity

Men 55-64, 2007

100

60 Other reasons No jobs

50 Family/care/personal reasons Illness/disability

40 Retired 30

gr fr it lv lt lu t nl at pl pt ro si fi 27 be bg cz dk de ee es cy hu m sk se eu Source: Eurostat, Labour Force Survey.

To conclude, the data presented in this section confirm the potential keeping the ageing babyboomers active for longer than previous cohorts in their late fifties and early sixties. A rising trend in employment rates of people aged 55-64 has now been firmly established, due, in particular, to the increased labour force participation of women. The higher level of educational attainment of the baby-boomers can be expected to contribute to a further increase in employment rates.

However, further analysis is required of the health status of older workers and to find out whether enough is being done to update the skills of ageing workers in accordance with the current needs of the labour market. Moreover, caring obligations towards grandchildren or dependent adults are likely to continue to represent an obstacle to increased employment of women in their fifties and sixties: at this age, they may be expected to care for grandchildren and their ageing parents. Other obstacles may be related to the attitudes of employers, employees and society in general towards older workers. Finally, social protection systems will have to provide incentives for staying longer on the labour market. All these issues are being given further attention, notably within the framework of the Open Method of Coordination on social protection and social inclusion. Moreover, new datasets such as the European Health Interview Survey and a second round of SHARE data will allow more indepth analysis.

3.2. Senior citizens' involvement in unpaid work

With large cohorts now reaching their sixties, a growing number of people most of whom are in good physical and mental condition are retiring from work and wish to maintain or create social ties with others in the community. For many older people, retiring from work also means opportunities for developing new activities in the form of unpaid work and volunteering. Four key dimensions of unpaid work can be distinguished: formal volunteering, informal helping, caring, and home production/housework

. Studies from a variety of

countries such as the US, Germany, or Australia have shown that senior volunteering and caring contribute to society in economically significant ways. Indeed, the economic contribution of non-profit institutions is estimated at 5% of GDP, and volunteer time accounts for 25% of this figure

This section focuses on informal helping and caring and on formal volunteering of older people and looks at evidence showing how these forms of unpaid work are related to the individual's demographic, socio-economic, and health characteristics, as well as their level of education.

How active are older people?

After their retirement from the labour market, senior citizens may pursue a wide range of other activities. The 2006 EU-SILC special module on social participation

provides an

46 See Active ageing: study on social participation and voluntary involvement of older people carried out for the European Commission by Dr. Karsten Hank, et alt., June 2008. This section draws extensively on this study. 47 Johns Hopkins University Report Measuring Civil Society and Volunteering , September, 2007. www.jhu.edu/ccss , quoted in the European Parliament resolution of 22 April 2008 on the role of

volunteering in contributing to economic and social cohesion ( 2007/2149(INI) )

48 EU-SILC Statistics on Income and Living Conditions - is an EU-wide survey with effective sample size of 121,000 households. A special module in 2006 module asked questios about social participation,

insight into European citizens' participation in various types of activities in a wide range of organisations such as churches and religious organisations, political parties or trade unions, recreational groups and charitable organisations, or informally outside any organisational context.

About one quarter of the population aged 65 or over participates in the activities of churches or other religious organisations, a slightly higher proportion than in younger age groups (see Table 3.3). Participation is highest in Cyprus (87.2%), Poland (almost 69.5%) and Ireland (62.5%). By contrast, fewer than one older person in 20 participates in such activities in France (2.4%) and Hungary (4.3%). In some countries, church and religious activities seem to appeal much more to older than to younger people. This is the case in Spain, where the participation of older people is 22 percentage points higher than that of people aged 25-64, but also in Ireland (+14 points), Luxembourg (+13 points), Slovenia and Lithuania (+11 points).

Table 3.3: Participation in activities of churches or other religious organisations, in EU24, by age group, in 2006, in %

All By age group

16-24 25-64 65+ Total 20.5 18.2 19.7 24.5 AT 13.6 8.5 13.8 16.9

CY 87.3 85.8 87.8 87.1 CZ 5.9 3.0 5.4 10.6

DE 15.4 12.4 14.5 20.2 DK 11.3 7.9 10.7 15.8 EE 5.3 2.4 4.9 8.9

ES 17.5 7.9 14.1 35.3

FI 15.8 12.1 14.5 22.8 FR 1.4 0.6 1.3 2.3

GR 29.1 20.1 29.1 34.5 HU 3.5 2.3 3.6 4.3

IE 49.0 41.7 48.4 62.6 IT 19.1 17.3 18.4 21.8 LT 21.0 13.5 20.3 30.4 LU 33.9 23.9 32.6 46.9 LV 8.9 5.6 8.2 14.0

NL 44.5 39.5 43.5 53.1 PL 68.7 68.9 68.5 69.2 PT 43.0 36.7 42.9 47.5 SE 19.6 13.8 19.0 24.4 SI 22.7 19.4 20.8 31.9 SK 35.9 33.9 34.5 44.1 UK 10.3 5.5 9.9 15.3 Source: EU-SILC module on social participation, 2006.

Participation in the activities of political parties or trade unions is generally low across the EU: it exceeds 10% of the entire population aged 16 or over in only two (Denmark and Finland) of the 23 Member States covered and remains below five percent in a majority of countries (see Table 3.4). Older people are in many cases more engaged than young people (16-24), but with the exception of one country (Czech Republic), they are less active in political parties or trade unions than middle-aged people (25-64).

covering cultural participation (e.g. cinema, sporting events) and frequency of contacts with friends and relatives.

Table 3.4 - Participation in the activities of political parties or trade unions, in EU24, by age group, in 2006, in %

All By age group

16-24 25-64 65+ Total 4.2 2.2 5.0 3.0 AT 5.6 2.2 6.6 4.5 CY 8.3 6.7 8.9 7.3 CZ 2.5 0.7 2.7 3.4 DE 6.4 3.8 6.9 6.4

DK 12.9 13.4 13.9 8.2 EE 3.7 2.1 4.7 1.8 ES 3.7 1.1 4.8 1.5

FI 11.1 5.3 13.9 5.7 FR 2.7 0.4 3.5 1.6 GR 5.0 3.6 6.6 1.4 HU 3.2 1.1 4.0 1.9 IE 4.0 2.0 4.9 2.9 IT 4.0 2.7 4.9 2.1 LT 2.0 0.8 2.5 1.4 LU 4.7 2.3 5.5 2.9 LV 7.0 2.7 8.5 5.7 NL 4.3 3.5 4.5 4.3 PL 3.7 1.6 4.7 1.9 PT 2.8 2.5 3.2 1.5

SE 8.9 8.1 10.5 5.0 SI 5.3 1.9 6.3 4.1 SK 7.3 2.1 9.6 3.5 UK 2.4 1.9 2.7 1.9 Source: EU-SILC module on social participation, 2006.

Older people are less active than both young and middle-aged people in recreational groups and organisations (see Table 3.5). Just under one fifth of older people in the 24 countries covered are involved in such activities, compared to one in four young people aged 16-24. Again, there are marked differences across countries, with participation of older people ranging from 1.7% in Poland to 42.5% in the Netherlands. Countries in Central and Eastern Europe and in the South are those with the lowest levels of participation, whereas in the North and West of Europe, participation levels typically exceed one quarter of the population aged 65 or over.

Table 3.5: Participation in activities of recreational groups or organisations, in EU24, by age group, in 2006, in %

All By age group

16-24 25-64 65+ Total 20.4 25.0 20.2 18.3 AT 22.9 23.3 24.3 17.5 BE 33.0 42.5 32.0 29.7 CY 29.8 34.0 29.7 25.0 CZ 21.8 31.5 21.5 14.7 DE 21.3 29.0 20.0 20.6 DK 33.8 31.4 35.9 26.5 EE 14.6 21.8 14.6 9.3 ES 13.8 24.4 14.1 6.1

FI 38.3 42.6 40.0 29.6 FR 23.2 25.0 21.8 26.4 GR 8.2 16.2 8.1 4.1 HU 5.7 9.7 5.7 2.7

IE 35.1 38.4 36.0 25.7 IT 10.4 15.2 10.5 8.0 LT 6.7 15.6 5.6 2.5

LU 35.4 44.0 36.5 24.0 LV 3.9 5.8 3.5 3.4

NL 46.8 53.7 46.5 42.5 PL 5.9 11.8 5.4 1.7

PT 11.2 18.0 11.8 4.6

SE 37.1 42.5 39.2 28.7 SI 19.8 31.7 21.0 7.0 SK 19.5 39.7 19.8 6.0

UK 34.5 29.8 34.5 37.9 Source: EU-SILC module on social participation, 2006.

A more detailed picture on social participation and activity of older people can be obtained from the Survey of Health, Ageing and Retirement in Europe (SHARE

), which covers only

twelve Member States. Data from the latest wave (2006-2007) of SHARE show that 19% of people aged 50 or over were engaged in general social activities (in clubs and political or community-related organisations) in the month prior to the interview, another 13.5% of respondents provided informal help outside their families, 11% were involved in voluntary or charity work in the month preceding the interview, and 6% provided care to other adults, typically to other family members (see Table 3.6).

49 See: http://www.share-project.org/t3/share/index.php?id=73

Table 3.6: Participation in informal activities in 12 EU Member States in the month preceding the interview (% of population aged 50+)

Country Voluntary or charity work Informal Care to General Help other social adults activities

AT 9.2 17.6 7.7 20.1

DE 13.1 14.9 7.4 25.4

SE 21.0 39.3 10.1 29.5 NL 25.5 24.3 11.1 35.2 ES 2.6 3.2 2.8 8.5 IT 8.4 6.9 3.7 9.4

FR 15.4 21.1 8.4 25.9 DK 21.2 26.8 5.6 42.1 GR 2.3 7.0 4.7 9.7

CH 16.8 19.3 10.4 37.4 BE 16.9 24.0 10.0 27.9 CZ 3.0 14.7 7.0 17.0 PL 2.0 4.7 3.7 3.9 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Cross-country comparisons reveal that informal help provided by older citizens is more common in Sweden (39%), Denmark, the Netherlands, and Belgium (24-27%). These countries are followed by France, and Austria, with 18-21% of the population aged 50 or older providing help, while low proportions of informal helpers are found in Greece, Italy (both 7%), and Poland (5%) and a particularly low proportion in Spain (3%). Care-giving is most common in the Netherlands (11%) and least widespread in Spain (3%). The various forms of social activities covered in Table 3.6 seem to be closely correlated. The country rankings by the levels of engagement in these four different types of activity are very similar. The Netherlands, Sweden and Denmark have the highest levels of participation whereas in three Southern countries, Greece, Spain and Italy, and two Central and East European countries, Poland and the Czech Republic, participation levels are the lowest.

The SHARE results highlight, in particular, a significant contribution of older people to childcare. Forty-five percent of grandmothers and 40% of grandfathers had looked after one of their grandchildren during the 12 months prior to the survey (see Table 3.7). Somewhat unexpectedly, the lowest proportions of grandparents caring for grandchildren are found in Spain, with 34% of grandmothers and 28% of grandfathers involved in the care of grandchildren, whereas the highest prevalence of this activity is observed in Denmark, France, Belgium, and the Netherlands. In many countries, the proportion of grandmothers and grandfathers providing childcare was similar, although in Italy and the Czech Republic grandfathers were much less likely to be involved than grandmothers. The reverse was the case in Belgium.

When the frequency of care for grandchildren is examined more closely, a different picture emerges. Sweden and Denmark, but also Germany and the Czech Republic, exhibit comparatively low levels of regular childcare by grandparents (lloking after grandchildren almost weekly or more often), whereas the proportion of Greek and Italian, but also Belgian grandparents who provide almost weekly childcare is roughly twice as high as that found in the Scandinavian countries.

Table 3.7: Grandparent who provided childcare over the past 12 months/since the last interview (in %)

Country Provision of any grandchildcare Provision of regular grandchildcare (almost weekly or more often)

Grandfathers Grandmothers Grandfathers Grandmothers AT 38.6 39.3 20.5 22.1 DE 36.8 38.1 17.1 19.2 SE 51.1 51.2 12.8 15.7 NL 54.5 53.5 26.3 25.7 ES 28.2 33.7 16.9 20.8 IT 34.8 48.2 26.2 36.1 FR 48.7 54.3 19.3 20.7 DK 52.5 55.0 11.7 16.5 GR 41.9 47.1 30.7 34.3 CH 44.6 46.8 19.6 25.5 BE 58.2 53.7 33.3 32.0 CZ 34.7 43.4 16.7 20.4 PL 41.3 46.2 25.6 30.8

Total 40.2 44.9 20.9 24.8 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

The most recent SHARE data show that, on average, 11% of the population aged 50 or older across the 13 European countries covered took part in voluntary work in the month prior to the interview. Substantial cross-country variation is found, with the largest proportions of senior volunteering in the Netherlands (26%), Sweden and Denmark (21%), whereas the proportions of volunteers in Poland, Greece (both 2%), Spain and the Czech Republic (both 3%) are particularly low (see Table 3.6 and figure 3.11).

Figure 3.11: Involvement of the population 50+ in voluntary work, by country

Proportion of volunteers

(in % of population aged 50+)

NL DK SE BE CH FR DE AT IT CZ ES GR PL

Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Across the countries surveyed, about two thirds of senior citizens volunteered weekly or more often (see Table 3.8). In three countries the low prevalence of volunteering (Poland, Greece

and the Czech Republic) was compounded by a low frequency of participation in voluntary work.

Table 3.8: Frequency of volunteering in the month preceding the interview, in % of the active population aged 50+

Country Volunteer

Almost daily Almost Less often every week

AT 6.6 39.5 53.9

DE 14.0 50.3 35.7 SE 15.3 46.0 38.6 NL 15.4 60.7 23.9 ES 28.1 41.0 30.9 IT 25.2 35.4 39.4 FR 23.7 46.4 29.9 DK 17.0 49.8 33.2 GR 13.5 33.5 53.0 CH 13.5 52.0 34.4 BE 22.0 47.5 30.5 CZ 14.8 34.5 50.7 PL 9.8 28.9 61.4

Total 18.6 47.1 34.4 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Informal help was provided almost daily by 29% of helpers in the 13 countries covered, and 70% helped at least almost weekly (see Table 3.9). The intensity of informal help was particularly high in Italy (45% helping almost daily), Greece (41.6%) and Spain (only 10.4% helping less often than almost weekly). It was relatively low in the Netherlands and Denmark where few people tended to provide help almost daily.

Table 3.9: Frequency of informal helping in the month preceding the interview, in % of the active population aged 50+

Country Informal Help

Almost daily Almost Less often every week

AT 22.9 44.6 32.5 DE 33.0 38.6 28.5 SE 25.7 39.0 35.3 NL 12.6 49.0 38.3 ES 36.2 53.4 10.4 IT 45.0 12.6 42.5 FR 28.4 38.4 33.1 DK 18.5 34.2 47.3 GR 41.6 43.8 14.7 CH 22.5 51.2 26.3 BE 28.5 49.7 21.9 CZ 30.4 32.0 37.6 PL 35.4 42.5 22.1

Total 29.0 39.6 31.4 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Social participation and individual characteristics

Social participation differs not only across countries, but also depending on individual characteristics such as gender, age, education and the employment status. The SHARE data provide some insights into how these characteristics are linked to social participation.

(a) Gender

On average, 12% of men and 10% of women engaged in voluntary activities (see Table 3.10 and Figure 3.12). However, substantially higher proportions of male volunteers are found in Germany (16% men vs. 11% women), Sweden (23% men vs. 19% women) and France (17% men vs. 14% women), whereas in the Netherlands slightly more women (26%) than men (25%) volunteered.

Figure 3.12: Engagement in volunteering by gender, and by country

30 Proportion of volunteers, by gender, and country (in % ) 25

NL SE DK BE CH FR DE AT IT PL ES CZ GR Male Female

Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

The same observations can be made about involvement in informal help, with 16% of men and 12% of women aged 50+ declaring to have provided such help in the month prior to the interview. The largest gender gaps can be found in Denmark, Austria and the Netherlands, where a larger proportion of men are involved in informal help (6 to 11 percentage points, see Table 3.10). By contrast, women were more involved than men in providing care.

Table 3.10: Participation in informal activities by gender, in %

Country Volunteer Informal Help Care General social activities

Men Women Men Women Men Women Men Women AT 10.4 8.2 21.7 14.3 7.1 8.1 25.0 16.2

DE 15.9 10.7 17.5 12.7 7.0 7.8 30.0 21.6

SE 23.2 19.1 40.3 38.4 9.8 10.6 30.7 28.5 NL 25.2 25.7 27.5 21.4 9.8 12.3 36.0 34.5 ES 2.9 2.4 3.7 2.9 1.6 3.9 9.6 7.5

IT 8.3 8.4 7.1 6.7 3.0 4.2 13.0 6.4

FR 17.4 13.8 24.9 17.9 6.3 10.1 28.2 24.0 DK 22.6 19.9 32.4 21.8 4.4 6.7 41.7 42.5 GR 2.2 2.3 6.4 7.5 2.3 6.8 13.8 6.1

CH 17.5 16.1 19.2 19.4 8.5 12.0 42.4 33.3 BE 18.0 15.9 26.3 22.0 9.0 12.0 33.3 23.3 CZ. 2.9 3.0 16.4 13.4 5.1 8.5 20.0 14.5 PL 3.4 0.9 5.9 3.8 2.7 4.4 5.0 3.0

Total 12.2 9.6 15.5 11.9 5.2 7.1 22.2 16.4 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

(b) Age

In the Netherlands, Sweden, and Denmark, the three countries with the highest prevalence of volunteering, the proportion of volunteers is 3 to 4 percentage points higher in the age group 65-74 than among people aged 50-64. A similar increase with age can be observed in Belgium, France and Germany, albeit to a lesser extent. By contrast, a decline in volunteering can be seen between the age groups 50-64 and 65-74 in countries with the lowest level of engagement in voluntary work, and in Switzerland. From the age of 75, volunteering typically drops to below half the level reached by the younger age groups.

Figure 3.13: Engagement in senior volunteering by age class, and by country, in %

Proportion of senior volunteers, by age class, and country (in % )

NL DK SE CH BE FR DE AT IT ES CZ PL GR Volunteer 50-64 Volunteer 65-74 Volunteer 75+

Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

A strong age gradient can be observed with regard to informal help. This form of social activity is most prevalent in the age group 50-64 and drops to about one third of that level in the age group 75+ (see Table 3.11). A similar age profile can be observed with regard to care provision, whereas the level of involvement in general social activities tends to diminish only in the oldest age group.

Table 3.11: Participation in informal activities by age, in %

Country Volunteer Informal Help Care General social activities

50-64 65-74 75+ 50- 65-74 75+ 50- 65-74 75+ 50-64 65-74 75+ 64 64

AT 12.2 8.8 2.6 23.1 16.7 6.0 10.2 6.0 3.9 24.2 21.2 9.3

DE 14.8 16.4 5.8 18.3 16.4 6.2 9.6 7.1 3.5 29.5 27.6 14.7

SE 21.0 25.5 16.1 50.5 38.2 14.1 12.6 8.6 6.3 31.4 32.0 22.2

NL 27.1 31.3 13.2 29.7 20.6 11.4 12.5 12.2 5.3 39.2 36.1 22.2 ES 3.7 2.0 1.1 4.3 3.4 1.1 3.9 2.8 0.8 12.0 6.8 3.5

IT 10.6 7.8 4.7 9.7 6.2 2.2 5.7 2.4 1.3 11.4 9.9 4.9

FR 16.9 17.5 8.6 24.8 20.2 10.3 10.3 7.1 4.0 26.5 28.9 20.3 DK 21.8 24.8 15.0 33.3 24.3 11.4 7.2 3.3 3.7 44.9 47.2 28.1 GR 2.6 2.3 1.5 8.5 7.0 3.1 5.9 4.2 2.6 11.5 9.6 5.0

CH 19.7 16.7 10.2 20.5 18.1 13.4 11.8 9.9 7.8 43.3 37.5 23.8 BE 18.3 19.0 11.0 29.6 23.3 10.8 11.9 9.4 5.8 30.1 28.4 21.8 CZ 3.7 1.9 1.7 16.6 13.9 8.8 8.3 4.9 5.0 19.0 16.3 10.2 PL 2.7 1.4 0.5 6.8 3.5 0.3 5.0 2.5 1.5 5.0 3.2 1.5

Total 12.4 11.9 5.5 17.2 13.0 5.5 8.2 5.4 2.9 21.8 19.8 11.5 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Older people with a higher level of education are more engaged in volunteering than those with a lower level of education (see Table 3.12 and Figure 3.14). Across the 13 countries surveyed by SHARE, participation rates for those with low formal education (5%) are 6 percentage points lower than for those with a medium level of formal education (11%), and 12 percentage points lower than for those with a high level of formal education (18%). Nevertheless, the poorly educated in the Netherlands (18%), Sweden (16%), and Denmark (15%) tend to engage more in volunteering than the highly educated in Italy (13%), Austria (9%), Spain (7%), the Czech Republic (6%), Poland (5%) and Greece (3%). Clearly, national culture remains a stronger determinant of volunteering than the individual level of educational attainment.

Figure 3.14: Engagement in senior volunteering by educational level, and by country, in %

35 Proportion of senior volunteers by educational level, and country (in % ) 30

NL BE SE DK FR CH DE IT AT ES CZ PL GR 8 9-12 years 13

Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Those with a higher level of education are also more often involved in informal help (19%, see Table 3.12), closely followed by those who have attained a medium level of education (16%), leaving those with the lowest level of education far behind (7%). Again, cross-country differences in the provision of help are in some instances more pronounced than differences between educational groups. In Sweden (25%) and France (18%), the proportion of older citizens with low formal educational achievement involved in informal help is larger than for those with higher educational levels involved in Spain, Italy, Greece and Poland. A higher level of educational attainment is also positively correlated with care-giving and involvement in general social activities.

Table 3.12: Participation in informal activities by level of education, in %

Country Volunteer Informal Help Care General social activities

8 8 8 8 9-12 13 8 8 8 8 9-12 13 8 8 8 8 9-12 13 8 8 8 8 9-12 13 years years years years

AT 8.0 11.7 9.3 14.9 25.3 16.5 8.5 7.3 6.2 18.2 24.9 19.9 DE 7.4 9.5 19.5 6.1 14.7 18.1 3.8 6.9 9.2 14.7 22.8 32.3

SE 15.8 22.8 23.9 24.5 43.1 48.5 6.6 11.1 12.5 22.0 29.5 36.1 NL 18.3 27.2 28.8 15.9 25.2 29.8 7.1 11.0 14.8 25.8 38.2 44.9 ES 1.3 4.1 7.1 2.4 4.1 6.3 2.2 3.1 5.6 6.7 10.7 14.2

IT 5.3 18.0 13.0 5.0 10.6 11.1 2.6 7.7 5.0 5.5 15.9 18.7

FR 8.8 16.0 22.2 17.8 23.2 21.8 4.7 9.0 11.5 19.9 24.8 33.5 DK 15.2 20.0 23.1 13.5 28.2 30.1 3.9 5.0 6.2 29.2 38.1 46.6 GR 1.9 2.5 3.0 5.8 6.9 11.7 3.8 4.3 9.0 7.4 9.4 18.8

CH 15.1 13.5 21.4 16.3 19.8 21.0 6.7 11.6 12.0 33.7 34.7 43.0 BE 9.1 14.3 24.7 15.2 22.4 31.4 6.5 8.7 13.6 19.6 26.5 34.7 CZ. 1.0 1.3 5.7 6.4 12.9 19.2 4.6 6.9 7.6 13.0 13.9 22.0 PL 0.8 2.7 4.5 1.5 6.3 11.9 2.1 4.3 7.7 0.9 5.8 9.4

Total 5.3 11.3 17.6 7.2 16.2 19.2 3.4 7.3 9.1 10.0 21.0 29.3 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

(d) Employment status

Across countries, the proportion of volunteers aged 50 or over differs only moderately between working (13%), retired (9%), and other non-working

(8%) people (see Table 3.14

and Figure 3.15), except in Austria, the Czech Republic (both -4 percentage points), and Poland (-3 percentage points), where volunteering is substantially lower among retirees than among those still on the labour market. In the Netherlands, the proportion of volunteers is higher among retirees and other non-working people than among the older working population, and the same is true in Belgium, although to a lesser degree.

Figure 3.15: Engagement in senior volunteering by employment status, and by country, in %

Proportion of volunteers by employment status, and country (in % )

SE NL DK CH FR BE DE AT IT CZ PL ES GR Working Retired Other

Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

Across countries, the proportion of senior citizens involved in informal help is larger among those who are working than among the two non-working groups. However, in Sweden and Denmark, the proportion of people in work aged 50+ involved in informal help is only marginally larger than that of non-working people who informally help others.

Participation in general social activities, on the other hand, is clearly linked with activity on the labour market, as the proportion of those still working who are taking part in social activities is 8 percentage points higher than among retirees, and 14 percentage points higher than other non-working people.

50 That is the unemployed and house people.

Table 3.13: Participation in informal activities by employment status, in %

Country Volunteer Informal help Care General social activities

Working Retired Other Working Retired Other Working Retired Other Working Retired Other AT 12.8 8.7 6.7 26.9 16.2 11.9 10.8 6.7 6.6 25.2 20.0 14.7

DE 15.3 11.8 14.1 18.8 13.0 15.0 9.2 5.9 9.6 33.4 23.5 20.0

SE 23.0 20.4 12.9 49.9 30.7 49.0 13.3 7.7 13.3 34.3 26.6 23.4 NL 21.7 27.4 27.1 30.3 21.0 21.9 12.7 9.7 11.2 41.4 31.0 33.8 ES 3.5 1.9 2.8 5.1 2.9 2.6 1.3 1.9 4.5 13.2 7.2 7.2

IT 10.6 9.3 5.0 9.5 7.3 4.3 6.4 2.7 3.7 13.3 11.1 3.2

FR 16.0 15.6 13.8 25.2 18.5 21.1 9.9 6.2 12.1 27.6 27.1 19.2 DK 21.3 21.3 19.9 34.0 20.8 29.4 6.4 4.0 10.0 45.6 41.8 30.8 GR 3.3 1.9 1.9 7.6 6.3 7.4 4.9 3.7 6.2 13.7 9.7 5.8

CH 19.7 14.3 16.7 22.4 17.8 16.3 10.6 8.7 14.8 45.3 32.7 32.0 BE 15.9 17.6 16.3 29.9 13.2 8.4 10.8 9.1 10.8 32.8 28.5 22.3 CZ 5.6 1.9 1.1 19.2 13.2 8.4 6.7 6.4 15.1 23.6 14.1 14.2 PL 4.7 1.4 1.6 8.5 3.4 5.5 5.9 3.1 3.7 8.3 3.2 2.5

Total 13.2 10.4 8.9 19.3 12.0 10.6 8.2 5.0 7.2 26.1 18.4 12.6 Source: SHARE 2006-07 (pre-release); calculations by the Mannheim Research Institute for the Economics of Ageing.

The variables considered here in relation to various forms of social activity are interconnected, and this needs to be taken into account when interpreting the findings. In particular, being very old makes it difficult to engage in such activities. There are more very old women than men, and this leads to a bias in the comparison by gender. Likewise, the very old will be over-represented among the retired compared to the population still at work. Correcting for the age structure would reduce the differences in social activities by employment status.

Cross-country differences in social activity dominate, however, over the differences linked to individual characteristics. The key challenge is, therefore, to gain a better understanding of those differences, and to see whether lessons can be drawn for policies to promote social activity and volunteering among the increasing number of older people in the EU.

Box 3.1: Eurobarometer results on volunteering

The Flash Eurobarometer Survey 247 asked respondents about their interest and engagement

in volunteering. Three-quarters of respondents who were not yet retired (73%) said they would consider participating in community or volunteer work. The proportion considering participating in community work or volunteering after their retirement ranged from 28% in the Czech Republic to 89% in Ireland.

Looking at retired people, one-third (34%) said they had participated in community or volunteer work since their retirement, while an additional 10% said they were planning to do

Since you retired, have you already or are you planning to...?

Participat(ing/ed) in community work or volunteering

Base: Retired respondents, % 'Yes, done this' & 'Yes, plan to do this'

6 11

50 9 11 7

7 24

11 12 10 13 40

10 13 9

30 15 15 17 7 19 10

50 8 9 10 13 45 45 12 44

20 38 11 37 37 13 35 34 11 33 32 28

23 22 22

10 21 21 20 20 19 18 17

12 12 0

NL FI LU DE FR UK IE AT SE SL EU27 DK IT BE MT HU LT PL CY LV EE ES GR PT RO SK CZ BG Yes, done this Yes, planning to do this so. Source: Flash Eurobarometer 247

Six out of ten Dutch and Finnish retired respondents said they had already participated or were planning to participate in community or voluntary work. In sharp contrast, only a quarter of the retired respondents in Bulgaria (23%), the Czech Republic and Slovakia (both 25%) had participated in voluntary work or were planning to do so.

3.3. Responding to the needs of an ageing population: policy implications 52

The Commission's Renewed Social Agenda underlined that Europe's ageing society

demands a variety of policy responses. A first priority for policy makers in the Member States must be to create more and better opportunities for active participation on the labour market and in society of the rapidly rising number of people in their sixties. Another key task for public policies is to ensure that the rising number of older people have adequate incomes as well as access to the goods and services that allow them to live independently for as long as possible. Finally, the ageing of the baby boomers requires strengthening solidarity with the increasing number of people in need of long-term care. The challenge for policy makers will be to provide sufficient care of high quality, so as to protect the physical and mental integrity.

All these policy responses fall within the responsibility of the Member States. However, the European Union can support their endeavours through mutual learning, monitoring of progress, and the definition of common objectives and targets. This is being done within the framework of the Lisbon Strategy and the Open Method of Coordination on social protection and social inclusion.

3.3.1. Policies to promote active participation in employment and society

The elements for a comprehensive strategy to promote the employment of older workers have been defined, and are being monitored, within the Lisbon Strategy. They include notably the improvement of working conditions and their adaptation to the health status and needs of older workers, better access to training and life-long learning, better access to Information and Communication Technologies (ICT)

and the review of tax-benefit systems to ensure that

there are sufficient rewards for working longer.

There also appears to be increasing readiness among policy-makers to promote voluntary work by older people. Such initiatives need to take into account the fact that new cohorts of older volunteers will tend to have higher levels of educational attainment, and more skilled professional backgrounds. This could allow them to make an effective contribution as volunteers, provided the right framework for mobilising their potential is put in place.

Box 3.2: The EU's contribution to active ageing in employment and society

· Within the European Social Fund (ESF), 1.01 billion euros are being devoted to measures

to promote active ageing and a longer working life for the programming period 2007-13. Older workers can also benefit from active and preventing labour market measures which represent 15.3% of ESF resources.

· The EU has established a general framework for equal treatment in employment and occupation (Council Directive 2000/78/EC of 27 November 2000) which also prohibits discrimination on the ground of age.

52 Communication from the Commission Renewed Social Agenda: Opportunities, Access and Solidarity in 21st Century Europe, 2 July 2008, COM(2008) 412.

53 SeniorWatch II report; an "Assessment of the Senior Market for ICT Progress and Developments". http://ec.europa.eu/information_society/newsroom/cf/itemdetail.cfm?item_id=4286

· Progress towards the common employment target of ensuring that 50% of people aged 55-

64 will be in employment by 2010 is being monitored within the Lisbon Strategy. The Social Protection Committee will focus on obstacles to more labour force participation of older workers in the 2009 Joint Report on Social Protection and Social Inclusion, which will comprise a chapter on 'Working more and for longer' .

· The European Council (Resolution on opportunities and challenges of demographic

change in Europe: the contribution of older people to economic and social development of February 2007) and the European Parliament (Resolution of 22 April 2008 on the role of volunteering in contributing to economic and social cohesion ) invited the Commission to take a more active role with regard to senior volunteering. Following the ENEA preparatory action on active ageing and mobility of older people, the Grundtvig programme for adult education will foresee the possibility of supporting senior volunteering.

· The Socio-economic Sciences and Humanities Programme of DG RTD finances a project

called "Activating Senior Potential in Ageing Europe" (ASPA) which will provide a comprehensive examination of the forces and mechanisms behind employers' decisions and public policies in relation to the use of senior potential in Europe, including human capital investments over the life course.

· The recent Commission proposal for a directive on equal treatment between persons

irrespective of religion or belief, disability, age or sexual orientation (COM(2008) 426 of 2 July 2008) could contribute to a better accessibility of goods and services for the disabled and frail elderly.

3.3.2. Policies to preserve the autonomy of older people

The autonomy and quality of life of older people depends not only on adequate retirement incomes, but also on access to a range of goods and services in areas such as personal finance, health, housing, built environment, communication and transport. Pension systems in the Member State currently offer most older people a high degree of financial autonomy and security

. However, 21% of women aged 65 and above have an income below the at-risk-of-

poverty threshold, compared to 16% of men in the same age group .

The reforms of public pension schemes leave more room for private pensions and other financial services in securing financial autonomy in old age. New financial products such as annuities and equity release schemes could make it easier for older households to convert wealth, particularly housing wealth, into a regular retirement income. However, the development of such products requires better financial education and more transparency.

Another key determinant of the autonomy and quality of life of older people is health. Physical and mental impairments can be prevented to some extent through healthier life styles

54 Minimum Incomes and Older Women's Poverty, 54 Minimum Incomes and Older Women's Poverty, http://ec.europa.eu/employment_social/spsi , 21 juin 2007., 21 juin 2007.

55 See Commission Staff Working Document Monitoring progress towards the objectives of the European Strategy

for Social Protection and Social Inclusion , 6 October 2008 ( http://ec.europa.eu/employment_social/spsi/docs/social_inclusion/2008/omc_monitoring_en.p df )

even if only adopted at a later age. Health care services need to be adapted so that they can respond to the specific health problems of older people and compensate for their disabilities.

However, most older people will have to cope at some point with health impairments. Whether this leads to more or less complete dependency will very much depend on the environment in which these older people live and on the services that are available to them from home. Homes and the local environments, including public transport, can be adapted and goods and services made more accessible so that more people with health impairments can continue to live in their own homes, maintain social contacts and look after themselves with a minimum of help from others. New technologies can play a major part in extending the possibilities of frail older people being able to cope in their daily life, and get help, when necessary.

Older people are the main users of long-term care, and their demand for such services can be expected to grow significantly over the coming decades. Member States will have to develop a coordinated supply of health and long-term care services adapted to the specific needs of older people, who often suffer from multiple and chronic health problems. In addition to the problem of funding such services, many Member States are also confronted with labour shortages in this sector. Low wages and poor recognition of professional carers result in high turnover and recruitment difficulties.

Frail older people are a highly vulnerable group and in view of the projected increase in the number of older people, it is crucial to address the question of safeguarding their fundamental rights and ensuring they are not exposed to the risk of neglect or abuse. Achieving this goal requires adequate provision of professional care, as well as support to family carers at present mostly women who, across the EU, bear most of the burden of care provision.

Meeting the specific needs of an increasing number of older people should not be seen as a burden, though. An independent study conducted for the Commission showed that a paradigm shift towards community-based care ("de-institutionalisation"), allowing older people to stay longer in their own homes, can represent an increase in quality without entailing higher costs

. The goods and services that are required for preserving the autonomy and quality of

life of an increasing number of older people represent a considerable economic opportunity. Independent living technologies can be expected to become a global growth market.

More and more people will experience old age outside their country of origin. They may have arrived in their host country as migrant workers, or they may have chosen to retire to another country, typically in the South of Europe. These elderly migrants, whose number can be expected to grow fast, often have specific needs which also need to be taken into account.

Mansell J, Knapp M, Beadle-Brown J and Beecham, J (2007) Deinstitutionalisation and community living outcomes and costs: report of a European Study . Canterbury: Tizard Centre, University of Kent (available for download on http://ec.europa.eu/employment_social/index/7002_en.html).

Box 3.3: The EU's contribution to autonomous living and long term care of older people

· The Social Protection Committee monitors in particular the adequacy of pensions. A

special report to be published in 2009 will look at the contribution of private pension schemes to the overall adequacy and sustainability of pension systems. The Commission has ordered a study on equity release schemes, as announced in the White Paper on the integration of EU Mortgage Credit Markets

· Regarding better access to basic financial services for disadvantaged groups the Commission intends to launch a public consultation on how to best prevent financial exclusion before the end of 2008.

· A new strategy for health has been adopted in October 2007 (Together for Health: A Strategic Approach for the EU 2008-2013, COM(2007) 630). One of its objectives is to foster good health in an ageing Europe.

· The 7 Framework Programme for research and development devotes 6.05 billion to

better health over the live cycle and, in particular, to the specific health problems of older people. The results of research funded under this programmes are expected to contribute to a better prevention of physical and mental impairment. 400 million will also be devoted to Information and Communication Technologies providing solutions for Telecare, independent living and mobility of elderly people.

· The Commission has opened a debate on urban mobility with its Green paper Towards a

new culture for urban mobility of September 2007 (COM(2007) 551). This should lead to an action plan which will also address the issue of accessibility for disabled people.

· The Commission's Disability Action plan 2008-2009 (COM (2007) 738) has defined

accessibility for all to goods and services as a priority. The Commission has issued two standardisation mandates to the European standardisation organisation in order to develop accessibility standards for information and communication technologies to be used in public procurement procedures.

· The Commission has presented in July 2008 a proposal for a directive on equal treatment

between persons irrespective of religion or belief, disability, age or sexual orientation (COM(2008) 426). The proposal concerns notably protection against discrimination in access to goods and services.

· The EU adopted a communication on Ageing Well in the Information Society (COM(2007)

332)

in November 2007, which presents an action plan to accelerate the introduction of new technology-based solutions for ageing well at home, in the community and at work.

· The EU supports the Ambient Assisted Living joint research and development programme 58

undertaken by several Member States . The programme aims at enhancing the quality of life of older people through the use of new Information and Communication Technologies.

57 COM(2007) 807du 18.12.2007. 58

Decision 743/2008/EC of the European Parliament and of the Council of 9 July 2008 (Official Journal L 201 of 30.7.2008, p. 58).

· The Socio-economic Sciences and Humanities Programme finances a project on

demographic change and housing wealth (DEMHOW ) which will investigate the links between ageing populations and housing wealth across Member States and the role of housing in providing income security in old age.

· The new Member States (EU-12) can use money from the European Regional Development Fund to invest in the development of their social housing stock.

· Through the ENEA preparatory action on active ageing on mobility of older people, the

Commission supports a major project focusing on the promotion of active ageing and social, cultural and economic integration of migrant and minority ethnic elders (AAMEE).

4.

T ACKLING THE C HALLENGES OF D EMOGRAPHIC C HANGE : U PDATE The Commission Communication on "The demographic future of Europe from challenge to opportunity" adopted in October 2006

presented five key policy directions through which

the Member States can respond to the challenges of demographic change:

1.

Promoting "demographic renewal" in Europe: creating conditions that support Europeans in fulfilling their desire to have children, notably by facilitating the reconciliation of work, family and private life. 2. Promoting employment in Europe: ensuring that more jobs of better quality are created and that people can work longer, thus achieving a better balance between active and inactive people. 3. Promoting a more productive and dynamic Europe: boosting productivity growth by optimising skills at all ages, thus strengthening the economy's ability to meet the needs of an ageing population. 4. Receiving and integrating migrants in Europe: alleviating future labour shortages by attracting skilled and unskilled workers from abroad and fostering their integration.
5.

Ensuring the sustainability of public finances: consolidating budgets and reforming social protection systems so as to guarantee adequate social protection and public services in the future.

The Communication also announced that once every two years the Commission would assess the Union's state of preparedness for demographic change. The present chapter aims to provide key data for such an assessment in each of the five policy domains above. The same data are also presented in the country sheets in the annex.

Each Member State faces different demographic challenges and, depending on the socioeconomic and political context, has its own, very specific, set of opportunities for tackling these challenges. The purpose of this chapter, therefore, is to help policy makers in each Member State to understand where their own country is positioned in relation to the rest of the EU, to see where there is the greatest potential for action and possibly also to identify other Member States that may have developed policies from which lessons could be drawn.

The data presented here present a snapshot of the current situation across the EU. The emphasis is not on long-term trends, but on the specific position of each individual Member State vis-à-vis the challenges of demographic change. It is up to policy makers, researchers and stakeholders in the Member States to analyse the specificities of their country and to derive appropriate policy responses from their analysis.

Progress in the different policy areas which contribute to tackling demographic challenges is closely monitored at the European level in different frameworks: the Lisbon Strategy, the Open Method of Coordination on social protection and social inclusion, the Stability and Growth Pact, the Roadmap for equality between men and women and the European Alliance for Families. The principal added value of this chapter and the country sheets that follow is to

59 COM(2006)571

bring together, in one place, indicators from this wide range of policy areas and to recall how they are related to the EU's ability to respond to demographic challenges.

4.1. Improving the conditions for Europe's demographic renewal

Recent data on total fertility rates show a slight increase (see Chapter 1), but in a majority of Member States, the average number of children per women is estimated to be lower than 1.5. If fertility levels stay at this low level, shrinking populations and much more pronounced ageing would result over coming decades. For this reason, low fertility rates have become a concern for a number of governments.

Governments have no direct influence on the decisions of people to have children, but they can try to achieve a return to somewhat higher fertility rates by creating a more supportive environment for families. This may comprise financial benefits, services (including affordable and good quality childcare and housing) and leave/working time arrangements that enable a better reconciliation of paid work and family life. This section focuses on financial benefits and services.

Family policies place considerable emphasis on financial support, through the payment of benefit or tax allowances. Such measures compensate families to some extent for the costs involved in raising children. Families also benefit from free or reduced-price services (notably childcare). Comprehensive sets of internationally comparable data on the value of these various forms of support to families do not exist. However, Eurostat collects data on how much Member States spend directly to provide

financial support to households for bringing up children;

financial assistance to people who support relatives other than children;

and social services specifically designed to assist and protect families, particularly

children.

Figure 4.1 shows the percentage of GDP devoted to such expenditure in each Member State in the year 2005. The highest levels of spending can be observed in Denmark and Luxemburg, followed by Germany, Austria, Finland and Sweden. Low levels of spending (in relation to GDP) can be found in Southern and Central and East European Member States. The three countries with the highest level of spending devote three to four times more of their GDP to families than the countries with the lowest level of spending.

Figure 4.1: Family benefits in % of GDP, in 2005

4.0 4.0 3 best countries

3.5 3.5 3.0 3.0 2.5 2.5 EU-27 average

2.0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0

K LU E AT FI IE U Y SI K R Z O LV N L PT LT IT T PL D D SE FR H C BE SK U EE G C R BG ES M

Source: Eurostat, ESSPROS database.

Note: Many values are provisional. Data for Portugal are for 2004. EU-27 average is estimated.

The spending on family benefits comprises both benefits in cash and benefits in kind. Figure 4.2 shows the distribution of total spending across these two types of benefit. For the EU as a whole, about three-quarters of social protection spending for families and children is on cash benefits and one quarter on services (benefits in kind). The Nordic countries, Spain and the Netherlands are distinguished by a very large proportion of benefits in kind, albeit in relation to a low overall level of spending in the last two countries.

Figure 4.2: Family benefits in cash and in kind, 2005

Source: Eurostat - ESSPROS database.

Note: Many values are provisional. Data for Portugal are for 2004. No data on benefits in kind for Poland: EU27 average is estimated.

The most important service to families is the provision of high-quality and affordable childcare. In view of the importance of childcare for raising employment rates, the 2002

Barcelona European Council set common targets for the EU to be achieved by 2010: 33% of all children aged 0-2 and 90% of all children aged 3 to the compulsory schooling age should have access to formal childcare. Figures 4.3 and 4.4 present the most recent data on the progress made by EU Member States in achieving the Barcelona targets.

Figure 4.3: Formal childcare capacity for all children aged 0-2 in %, in 2006

100 100 90 90 80 80 70 70

60 3 best countries 60 50 50 40 40 EU-25 average

30 30 20 20 10 10 0 0

D K N L S E B E E S PT U K FR LU SI IT FI C Y D E E E IE LV G R H U M T S K LT AT C Z P L

Source: Eurostat, EU-SILC.

Notes: Data for Bulgaria and Romania are not available. 2006 data are provisional for BE, DE, EL, FR, IE, LT, LU, LV, MT, NL, PL, PT, SE, SK and UK. The age of children is calculated at the date of the interview, except for IE and FI where age is calculated at 31 December 2005. For CY, LV, PT and SK, no information was collected for children born between 31 December 2005 and the date of the interview.

Figure 4.4: Formal childcare capacity for all children aged 3 to compulsory school age in %, in 2006

100 3 best countries 100 90 EU-25 average 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0

BE K FR E IE SE ES IT N L K C Y EE SI U FI P T SK A T C Z R LV LU T LT P L D D U H G M

Source: Eurostat, EU-SILC. Notes: see above, Figure 4.3.

For both age groups, the EU is coming close to the target, but considerable differences can be observed across Member States. Most of the former communist Member States have very low levels of childcare provision, both for the youngest and older children. The country ranking differs between the two age groups. Denmark performs best with regard to childcare for children under the age of three. For the age group 3-6, a group of eight countries exceeds or

reaches the Barcelona target: Belgium followed by Denmark, France, Germany, Ireland, Sweden, Spain and Italy. The figures do not take account of the hours of childcare offered

When formal childcare can only be used for half a day, other arrangements are necessary for the rest of the day, implying that one parent may only able to work part time. Some informal childcare arrangements, such as the assistantes maternelles are also not included in these figures which, therefore, only provide an incomplete picture of the situation.

The development of childcare services is crucial for promoting the labour force participation of women. Figures 4.5 and 4.6 present the employment rates of women and men who are caring for at least one child below the age of six. Whereas close to 90% of men with at least one young child are in employment, the corresponding employment rate for women is less than 60%; in addition, a large proportion of women are working part-time (see Figure 4.7).

This clearly shows that it is still mainly women who adjust their employment situation to suit the needs of their families. The `male breadwinner' model seems to be going particularly strong in countries like Malta, the Czech Republic, Hungary and Slovakia, where women with young children have employment rates below 40%. Differences across countries are also much greater as far as the employment of mothers of young children is concerned than in the case of fathers. The spread between the country with the highest employment rate for women with young children and the country with the lowest rate is close to 50 percentage points, compared to around 20 for men.

Figure 4.5: Employment rate of women having at least one child below 6, in 2007, in %

80 Women aged 20-49 70

L Y T FI E K R IT T Z U SI PT LT LV A PL

N C BE LU FR ES D U EE G SK M C H

Source: Eurostat, Labour Force Survey. Note: Data for Denmark, Ireland and Sweden are not available.

60 See the Commission Working Document "Mise en oeuvre des objectifs de Barcelone concernant les structures d'accueil pour les enfants en âge préscolaire" SEC(2008)XXX.

Figure 4.6: Employment rate of men having at least one child below 6, in 2006, in %

100 100 3 best countries

EU-27 average

90 90 80 80 70 70 60 60 50 50 40 40

R CY N L LU PT IT FI CZ ES SI EE FR D E AT M T U K BE LV LT PL SK H U O BG G R

Source: Eurostat, Labour Force Survey. Note: Data for Denmark, Ireland and Sweden are not available.

Figure 4.7: Proportion of part time workers among employed women, in 2007, in %

80 80 70 70 60 60 3 highest

50 50 40 40 EU-27 average

30 30 20 20 10 10 0 0

L E K T E E K IE IT T S FI T L E S I Y O LT R Z LV U K G N D U A B S LU D FR M E P P E C R G C H S B

Source: Eurostat, Labour Force Survey. Note: Data for Ireland are from 2004.

The fact that women tend to adapt their labour market involvement to the needs of their families is also likely to be a key factor in the large pay gap between women and men (see Figure 4.8). The gender pay gap is the difference between average gross hourly earnings of male and of female paid employees as a percentage of average gross hourly earnings of male paid employees. The population considered consists of all paid employees aged 16-64 who work at least 15 hours per week. Across the EU, women earn around 15% less than men. The pay gap is 20% or higher in Estonia, Cyprus, Germany, Slovakia, the UK, Austria and Finland. Eight countries have a pay gap of 10% or less, and the small numbers of women in employment in Malta enjoy the highest level of pay in relation to men.

Figure 4.8: The Gender Pay Gap in %, in 2006

30 30 25 25 20 20 EU-27 average

15 15 10 10 3 best countries

5 5 0 0

EE Y E SK K AT FI N L C Z K LV LT SE LU ES PL FR U R IT IE PT SI BE T C D U D BG H G RO M

Source: Eurostat, EU SILC and national sources. Note: Data for DK, DE, EE, IT, LT, NL, PT and UK are from 2005.

The data presented above suggest that there is considerable room for improvement in many Member States and in different areas of family policy and gender equality (see also Chapter 2 on the policy implications of changing family structures). Choosing the right policy mix is crucial in responding to the challenge of very low birth rates, and the data presented here can provide only a few indications. Further, more in-depth work will be carried out in the framework of the European Alliance for Families.

4.2. More employment

The main indicator used to describe the ageing of a society is the old-age dependency ratio, which divides the number of people aged 65+ by the working age population (aged 15-64). In 2008, the ratio stood at 1 older person for 4 people of working age. It is expected to rise to 1 for 2 over the next 40 years. However, the ability of a society to cope with an ageing population does not simply depend on the ratio between these two age groups. The key question is how many inactive people, and people with expensive health and long-term care needs, have to be supported by the active population.

The active population is in fact much smaller than the age group 15-64. A very large proportion of young people under the age of 25 are still in education or training, while most people retire well before they reach the age of 65. Among those in between, aged 25-59, many are not in employment: a significant proportion of women, for family reasons, and a large proportion of women and men with a low level of educational attainment. This leaves considerable scope for increased employment in most Member States and, consequently, an opportunity for achieving a much more favourable balance between the population in employment and retired older people. Indeed, the 2006 Demography Report

estimated that

raising the EU employment rate to the level of the three best-performing Member States could compensate for about two-thirds of the decline in employment expected as a result of the shrinking of the working-age population. This illustrates the importance of raising

61 European Commission, Directorate-General for Employment, Social Affairs and Equal Opportunities: Europe's Demographic Future: Facts and Figures on challenges and opportunities. SEC(2007)638

employment levels in the EU. It is arguably the most effective strategy with which countries can prepare for population ageing.

Achieving higher levels of employment is also at the core of the Lisbon Strategy, which set ambitious goals in this regard, namely to raise the total employment rate to 70% by 2010. By 2007, 7 Member States had reached this goal (see Figure 4.9): Denmark, the Netherlands, Sweden, Austria, the UK, Cyprus and Finland. Germany, Estonia and Ireland were very close to the target figure. The three best performing countries demonstrate that an employment rate of 75% is possible.

The EU average is still below the 70% target, at 65.4%, which represents a growth of almost 5 percentage points compared to the level of 60.7% in 1997, but significant differences exist across countries. The Northern and Western European countries all have rates above the EU average, whereas the Mediterranean (Malta, Italy, and Greece) and Central and East European countries (Poland, Hungary, Romania, Slovakia) tend to have the lowest employment rates.

Figure 4.9: Total employment rate of persons aged 15-64 in %, in 2007

80 80 3 best countries

75 75 70 70 EU-27 average

65 65 60 60 55 55 50 50 45 45 40 40

K N L SE AT K Y FI E EE IE LV PT SI C Z ES LT FR LU BE R SK O IT U PL T D U C D BG G R H M

Source: Eurostat, Labour Force Survey.

Figure 4.10: Employment rate of women aged 15-64 in %, in 2007

80 80

75 3 best countries 75 70 70 65 65 EU-27 average

60 60 55 55 50 50 45 45 40 40 35 35 30 30

K N L FI K LV AT E SI Y LT PT IE Z LU O U PL R IT T D SE EE U D C FR BG C BE ES SK R H G M

Source: Eurostat, Labour Force Survey.

Figure 4.11: Employment rate of men aged 15-64 in %, in 2007

90 90 85 85 3 best countries

80 80

75 EU-27 average 75 70 70 65 65 60 60 55 55 50 50 45 45

N L K Y AT IE K R Z E T PT SI LV FI LU IT LT O U PL D C U SE ES G C D M EE FR BE SK BG R H

Source: Eurostat, Labour Force Survey.

Differences in total employment rates reflect to a large extent differences in employment rates of women (see Figure 4.10). The Lisbon Strategy also sets a target of a 60% employment rate for women, a level that, in 2007, had almost been reached by the EU as a whole and by 15 Member States. The countries that have yet to reach this target are from Southern, Central and Eastern Europe, apart from Belgium and Luxembourg. The best-performing countries achieve a female employment rate of just over 70%, around 10 percentage points below the average of the three highest employment rates for men.

Eight Member States had male employment rates below 70%: France, Belgium and six new Member States (see Figure 4.11). Achieving the Lisbon employment target also requires determined efforts to raise the labour force participation of men, notably by helping older workers stay on the labour market (see Chapter 3).

4.3. Higher productivity

It is not only the number of jobs that determines a country's prosperity, but also the quality of jobs, for which labour productivity can be used as a proxy. Large differences are found in productivity levels across the EU. Figure 4.12 shows that the highest hourly productivity levels expressed in Purchasing Power Standards are in the Benelux countries and France, at around 120% of the EU-15 average. The best performing countries produce around four times as much output per hour worked as the poor performers, Bulgaria and Romania. All the Member States that joined the EU in 2004 are significantly below the EU average of around 90. If productivity is measured in Euros, the differences are even larger.

Figure 4.12: Labour productivity in GDP per hour worked in PPS, EU-15=100, in 2006

180 180 160 160 3 best countries

140 140 120 120

100 EU-27 average 100 80 80 60 60 40 40 20 20 0 0

LU N L E IE K AT FI K IT T SI R Y PT U Z PL LT LV O BE FR D SE D ES U M G C SK H C EE BG R

Source: Eurostat - National Accounts. Data for Romania are from 2005.

The productivity level achieved in a country reflects the level of technology and past investments in physical and human capital (including the health of workers). Investment in human capital, notably through education and training, plays a crucial role in this process. This section focuses on investment in human capital and looks at differences in educational attainment of the working-age population across Member States. The need to increase investment in human capital through better education and skills is fully recognised in the Employment Guidelines 2008-2010. The EU has set itself ambitious quantitative targets in this area which are to be reached by 2010: not more than 10 % youths should leave schools early and 85 % of the 22-year olds should have completed upper secondary education. Moreover, the average level of participation in lifelong learning should reach at least 12.5 % of the population aged 25-64

Figures 4.13 and 4.14 present the proportion of early school leavers for both sexes in 2007, defined as young people aged 18-24 with at most secondary education and not in further education or training. The EU-27 average for women lies at 13% and for men at 17% in 2007. The European Benchmark for early-school-leaving was put at no more than 10%. The gap between the EU-27 average and the three best performers for both genders amounts to about 10 percentage points. The largest proportion of early school leavers are found in Portugal, Spain and Malta with rates above 25% and 35% for women and men respectively.

For a more detailed analysis see "Future skill needs in Europe: Focus on 2020", European Centre for Development and Training (CEDEFOP), 2008, http://www.cedefop.europa.eu/index.asp?section=3&read=3650

Figure 4.13: Early school-leavers, % of the women aged 18-24, with at most lower secondary education and not in further education or training, in 2007

40 40 35 35 30 30 25 25 20 20

15 EU-27 average 15 10 10 3 best countries

5 5 0 0

T PT O IT LV E K LU R AT N L U K IE Y FI LT Z PL SI M ES R BG D U FR EE BE G SE H D C SK C

Source: Eurostat, Labour Force Survey. Figures for CZ, SE, UK are from 2006, figures for EE are from 2005. Figures for LT, LU, SL should be regarded as unreliable.

Figure 4.14: Early school-leavers, % of men aged 18-24 with at most lower secondary education and not in further education or training, in 2007

45 45 40 40 35 35 30 30 25 25

20 EU-27 average 20 15 15 10 10 3 best countries

5 5 0 0

PT T IT LV Y LU O R K K N L IE E U AT LT FI PL Z SI M ES EE C R G BG D FR U BE D SE H SK C

Source: Eurostat, Labour Force Survey.

The same three best performing countries are also distinguished by a small proportion of young people who have completed at least upper secondary education (see Figures 4.15 and 4.16), followed by Germany. The adopted European Benchmark says that at least 85% of young people should have completed upper secondary education. The best performing countries, with regard to both early school leaving and attaining at least upper secondary education, are the Central and East European Member States: Slovenia, the Czech Republic, Poland, Slovakia and Lithuania; Finland also displays a low rate for early school leaving.

Figure 4.15: Youth education attainment level: % of women aged 20-24 having completed at least upper secondary education, in 2007

100 100 3 best countries

90 90 EU-27 average

80 80 70 70 60 60 50 50

SI PL C Z SK LT C Y IE EE SE FI R HU AT FR BE LV NL IT UK DK O LU DE ES PT M T G BG R

Source: Eurostat, Labour Force Survey.

Figure 4.16: Youth education attainment level: % of men aged 20 to 24 having completed at least upper secondary education, in 2007

95 95 3 best countries

90 90 85 85 80 80 EU-27 average

75 75 70 70 65 65 60 60 55 55 50 50 45 45

C Z SK PL SI LT SE FI IE AT U BE FR C Y R U K LV IT EE N L D E LU D K ES T PT BG H G RO M

Source: Eurostat, Labour Force Survey.

Tertiary education is becoming increasingly important for competitive, knowledge-based economies. The number of university graduates in 2006 per 1000 people aged 20-29 is presented in Figure 4.17. Lithuania leads the ranking, followed by the UK, Denmark, the Netherlands and Poland. Germany, Austria and Latvia are found at the lower end of the scale.

Figure 4.17: University graduates aged 20-29 per 1000 persons of the corresponding age population both sexes, in 2006

70 70

60 3 best countries 60 50 50 40 40 30 30 20 20 10 10 0 0

LT U K D K N L P L BE EE FI O R PT T ES SI SK SE IT U C Z C Y D E AT LV R G M BG H

Source: Eurostat - joint UIS/OECD/Eurostat (UOE) data collection on education statistics using ISCED 5-6. Data for Ireland, France and Luxembourg, as well as the EU-27 average, are not available.

A high level of education not only enables workers to be more productive, it also increases their likelihood of being in employment. About 84% of people with tertiary education were employed in 2007, 70% of people with no more than upper secondary education and only 49% with only primary education.

The chances of finding, and remaining in, high-quality employment not only depend on the level attained during initial education, but also on keeping knowledge and skills up to date throughout working life. Participation in life-long learning is, however, still relatively rare in most Member States. Figure 4.18 shows that about 1 in 10 workers had taken part in some form of education or training over the four weeks prior to being surveyed, while the European Benchmark says that it should be 1 in 8. The level was up to three times as high in the best performing countries, whereas in the worst performing countries, workers hardly received any education or training at all.

Figure 4.18: Life-long learning, % population aged 25-64 participating in education and training over the four weeks prior to the survey, both genders, in 2007

40 40 35 35 30 30

25 3 best countries 25 20 20 15 15 EU-27 average

10 10 5 5 0 0

E K K FI L S I T S Y E IE E E IT T Z LT L T K U R G O S D U N A E C D FR B LV E LU M C P P S H G B R

Source: Eurostat, Labour Force Survey. Data for Sweden and the United Kingdom are for 2006.

Public spending on education also differs widely across the Member States, with Denmark spending more than 8% of its GDP in 2005 and Sweden and Cyprus around 7% (see Figure 4.19). At the other end of the spectrum, Malta spends under 3% and Romania about 3.5%. Luxembourg also devotes a relatively low percentage of GDP to education, but this is also a reflection of the high level of GDP per capita. No clear link appears between the proportion of GDP used for public spending on education and the outcome indicators (early school leavers, proportion of graduates) presented above. Thus, the quality of educational provision may be a more important factor than the amount spent.

Figure 4.19: Public spending on education as % of GDP, in 2005

10 10 9 9

8 3 best countries 8 7 7 6 6 EU-27 average

5 5 4 4 3 3 2 2 1 1 0 0

K Y FI SI PL U K L LV LT IE E IT Z R O T

D SE C BE FR H U AT PT N EE D BG C ES G SK LU R M

Source: Eurostat - joint UIS/OECD/Eurostat (UOE) data collection on education statistics.

Productivity growth is not only driven by increasing and improving fixed capital and human resources, but also by innovation. A combination of highly educated people and spending on research and development are prerequisites for adopting know-how developed elsewhere and for extending the technology frontier. Figure 4.20 presents expenditure on research and development in 2006 as a percentage of GDP. Sweden and Finland stand out with spending levels around 3.5% of GDP. They are followed by Germany, Austria and Denmark, but spending in these countries is about 1% of GDP lower than in the best performing countries. The EU's least developed countries also have the lowest levels of R&D spending at around one quarter of the EU average of 1.84% of GDP. This average is well below the target set for 2010 of three percent of GDP. Moreover, the level of R&D spending has not risen since the beginning of the decade.

Figure 4.20: Expenditure on R&D as% of GDP, in 2006

5.0 5.0 4.5 4.5 4.0 4.0

3.5 3 best countries 3.5 3.0 3.0 2.5 2.5 EU-27 average

2.0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0

SE FI E AT K FR BE K N L SI C Z LU IE ES EE IT U PT LT LV R PL T SK Y D D U H G M BG RO C

Source: Eurostat, WG on Statistics on Science, Technology and Innovation. Note: Data for Italy are for 2005.

4.4. Receiving and integrating migrants

The EU has been receiving an unprecedented number of migrants over recent years (see Chapter 1). In addition, many of the migrants who arrived over the past few decades have settled and raised their families in Europe. In many Member States, a significant proportion of children and young people have immigrant parents, and may have difficulty in integrating even if they are citizens of an EU Member State.

Figure 4.21 presents the proportion of non-nationals in each Member State. This only partly represents the scale of immigration since a proportion of immigrants may have received the citizenship of their host country. Apart from Latvia, Estonia and Cyprus, the countries with the highest proportion of non-nationals are EU-15 Member States, many of them counting between 5% and 10% of non-nationals among their populations. In the EU-12 Member States, the proportion of non-nationals tends to be significantly lower, with the exception of Latvia and Estonia, where so-called "recognised aliens", who have no citizenship of any existing country, Russian citizens, and citizens of other countries that became independent after the end of the USSR account for most of the non-nationals, and Cyprus where nearly six nonnationals out of ten come from another EU Member State.

Figure 4.21: Proportion of non-nationals in the EU-27 population, in %, in 2007

LU LV EE CY IE ES AT DE BE GR UK FR SE DK IT NL PT MT CZ SI FI HU LT SK BG PL RO

Source: Eurostat demographic data.

It is estimated that, over the past 13 years (1995-2007), the population of EU-27 increased by nearly 15.5 million people due to immigration, 4.5 million during the first 7 years and 11 million during the last 6 years of this period. The countries that attracted the largest numbers of migrants were Spain, Italy, Germany and the UK (see Figure 4.22). Six Member States lost population (Bulgaria, the three Baltic countries, Poland and Romania). Luxembourg stands out as the Member State with the largest percentage of foreigners.

Figure 4.22: Cumulated net migration (including corrections), 1995-2007

Cumulated net migration 1995-2007

s 6000 d

T 5000 4000

3000

2000

1000

be bg cz dk de ee ie gr es fx it cy lv lt lu hu mt nl at pl pt ro si sk fi se uk -1000

Source: Eurostat demographic data.

Apart from the steep increase in the immigration flow, there has also been a marked change in the main destinations countries of these flows. In particular, migration to Germany more than

halved, while it increased considerably to Spain and Italy, which have become the main receiving countries (see Figure 4.23).

Figure 4.23: Net migration flows (including corrections), annual averages 19952001 and 2002-2007

Net migration flows, annual averages 1995-2001 and 2002-2007

s 700 d

o 1995-2001 h 600 T

2002-2007

500

400

300

200

100

be bg cz dk de ee ie gr es fx it cy lv lt lu hu mt nl at pl pt ro si sk fi se uk -100

Source: Eurostat demographic data.

In relation to other receiving countries, Spain remains one of the countries that has absorbed the largest immigration flows, although it has now been overtaken by Cyprus, and Italy has been overtaken by Ireland. Crude rates of net migration also show more clearly the extent of emigration from some of the new Member States. These population losses were much smaller since 2002 than during the preceding period (see Figure 4.24).

Figure 4.24: Crude rates of net migration (including corrections), annual averages 1995-2001 and 2002-2007

Crude rates of net migration

20,00

1995-2001

2002-2007

t s

n 15,00 i t a

i n

10,00

r 1

) p

t i o

5,00

r r e

i n

l u

0,00

( i n

n ie fx it lv lt lu t nl pl si fi

27 be bg cz dk de ee gr es cy hu m at pt ro sk se uk t i o eu r a

i g

t m -5,00 e

-10,00

Source: Eurostat demographic data.

Attracting third country nationals is one way of preventing labour force shortages in a context of a declining working-age population. How successful such a strategy is depends, however, on the ability to integrate migrants into the labour market and allowing them to develop their full productive potential. Employment rates tend to be lower for men who are not nationals of an EU-27 Member State than for nationals of the country in which they live (see Figure 4.25). However, the situation differs considerably from one country to another. In the Czech Republic, Estonia, Greece, Spain, Italy, Lithuania, Hungary, Poland, Portugal, Romania, and Slovenia, third-country nationals are more likely to be in employment than nationals. Thirdcountry women are also less likely to be in employment than native women, and the gaps between third-country nationals and nationals are often even more pronounced than for men.

Figure 4.25: Comparison of employment rates of nationals and third-country nationals

Employment rates: nationals and non-EU27 foreigners

Men, aged 15-64, 2007

Nationals

Non-EU27

EU27 BE BG CZ DK DE EE IE GR ES FR IT CY LV LT LU HU MT NL AT PL PT RO SI SK FI SE UK

Employment rates: nationals and non-EU27 foreigners

Women, aged 15-64, 2007

Nationals

80 Non-EU27 70

EU27 BE BG CZ DK DE EE IE GR ES FR IT CY LV LT LU HU MT NL AT PL PT RO SI SK FI SE UK

Source: Eurostat, Labour Force Survey.

Note: data are missing for third-country nationals in Bulgaria, Ireland, Malta and Slovakia.

A higher level of education facilitates integration into the labour market and society. It is therefore interesting to look at levels of educational attainment of immigrants compared to that of nationals, particularly for countries with relatively large proportions of non-nationals.

Figure 4.26 shows that in several Member States with relatively large foreign populations, non-EU-27 nationals tend to have significantly lower levels of educational attainment than nationals. In Belgium, Germany, Greece, France and Italy, around half of these non-nationals have only received a low level of education. However, in Spain and the UK, two countries that have been receiving large numbers of immigrants over recent years, the difference between nationals and third-country nationals is small.

Figure 4.26: Comparison of educational attainment levels*, nationals vs. thirdcountry nationals Women

Men

0,0% 20,0% 40,0% 60,0% 80,0% 100,0% 0,0% 20,0% 40,0% 60,0% 80,0% 100,0% Nationals Nationals

E Non EU27 E Non EU27 B B

Nationals Nationals

G Non EU27 G Non EU27 B B

Nationals Nationals

Z Non EU27 Z Non EU27 C C

Nationals Nationals

E Non EU27 E Non EU27 D D

Nationals Nationals

E Non EU27 E E Non EU27 E

Nationals Nationals

R Non EU27 R Non EU27 G G

Nationals Nationals

S Non EU27 S Non EU27 E E

Nationals Nationals

R Non EU27 R Non EU27 F F

Nationals Nationals

I T Non EU27 I T Non EU27 Nationals Nationals

Y Non EU27 Y Non EU27 C C

Nationals Nationals

L V Non EU27 L V Non EU27 Nationals Nationals

T Non EU27 L T Non EU27 L

Nationals Nationals

L U Non EU27 L U Non EU27 Nationals Nationals

U Non EU27 U Non EU27 H H

Nationals Nationals

T Non EU27 T Non EU27 M M

Nationals Nationals

L Non EU27 L Non EU27 N N

Nationals Nationals

T Non EU27 T Non EU27 A A

Nationals Nationals

L Non EU27 L Non EU27 P P

Nationals Nationals

T Non EU27 T Non EU27 P P

Nationals Nationals

O Non EU27 O Non EU27 R R

Nationals Nationals I I

S Non EU27 S Non EU27 Nationals Nationals

K Non EU27 K Non EU27 S S

Nationals Nationals I I

F Non EU27 F Non EU27

K Nationals K Nationals U Non EU27 U Non EU27

Low Medium High Low Medium High Source: Eurostat, Labour Force Survey 2007.

*Low corresponds to ISCED 1,2,3c short, medium to ISCED 3a, 3b, 3c long , 4 and high to ISCED 5 and 6.

63 See also the research presented in the forthcoming 2008 Employment in Europe Report and in the impact assessment accompanying the proposal for a council directive on the conditions of entry and residence of third country nationals for the purpose of highly qualified employment (SEC(2007)1403 of 23.10.2007).

A key issue with regard to the integration of migrants is to ensure that their children are offered equal opportunities to develop their full potential. This remains a major challenge, as was shown in the 2007 Social Situation Report, which highlighted the fact that much larger proportions of children of migrant families grow up in poverty than do children of nationals

Moreover, the OECD's Programme for International Student Assessment (PISA) also highlights the difficulties children from a migration background are facing in EU education systems.

4.5. Sound public finances

The increase in the number of older people over the coming decades will create additional

public expenditure demands for pensions, health and long-term care. Reforms of social protection systems, making them more efficient and encouraging older workers to stay longer on the labour market, can curb the increase in expenditure to some extent. Governments can, however, also prepare for the needs of an ageing society by reducing their public debt and hence the amount of tax revenue they need to allocate for interest payments.

In 2007, government debt amounted to 60% of annual GDP in EU-27, the lowest level for

the past 12 years. Three countries are distinguished by more than 80, 90 and 100% debt to GDP ratios respectively: Belgium, Greece and Italy. Most of the new Member States have low debt burdens, 7 of them below 30% of their annual GDP (see Figure 4.27).

Figure 4.27: General government debt (% of GDP), 2007

General government debt (Percentage of GDP), 2007

110 110 100 100 90 90 80 80 70 70 EU-27 average

60 60 50 50 40 40 30 30 20 20 10 10 0 0

IT R E U E T T Y T L L K E S FI K Z K IE S I G O E G B H D FR P M C A N P U S E S C D B LT R LV LU E

Source: Eurostat.

The total amount of government debt is determined by annual budget deficits, which increase the debt ratio, and the rate of growth of nominal GDP, which decreases it. In 2007, the sum of budget deficits for EU Member States represented 1% of EU-27 GDP, down from 3% 5 years earlier. The situation varies, however, considerably across Member States and deficits ranged from a budget surplus of over 5% of GDP in Finland to a deficit of the same size in Hungary (see Figure 4.28).

64 European Commission, Directorate-General for Employment, Social Affairs and Equal Opportunities: The Social Situation in the European Union 2007 - Social Cohesion through Equal Opportunities.

Figure 4.28: Government surplus/deficit (Percentage of GDP), 2007

Government surplus/deficit (Percentage of GDP), 2007

6 6

5 3 best countries 5 4 4 3 3 2 2 1 1 0 0

-1 -1 EU-27 average

-2 -2 -3 -3 -4 -4 -5 -5 -6 -6

FI K E G Y E S L IE E S I E T Z T IT L K O T R K U D S B C LU E E N D LV B A LT C M P S R P FR G U H

Source: Eurostat.

The total amount of government debt, together with the level of interest rates, determines

the amount governments have to pay for interest on their public debt. Figure 4.29 shows that 6% of total public spending is used for interest payments. This amounts to just under 3% of GDP. Italy and Greece are obliged to use the highest proportion of its public spending to fund debt interest at 10%, followed by Hungary, Malta and Belgium at around 8% percent.

Figure 4.29: Proportion of public expenditure accounted for by debt interest, 2007

12 12 10 10 8 8 EU-27 average

6 6 4 4 2 2 0 0

IT R U T Y E PL AT NL K FI K SI Z IE LT O LV LU G H M BE C PT D FR U ES SK SE D BG C R EE

Source: Eurostat.

Debt volumes and deficit levels are one aspect of the assessment of the long-term sustainability of public finances and hence the ability of governments to meet the future needs of their ageing populations. In 2009, the Commission, in cooperation with the Economic

Policy Committee, will present a fresh assessment of the long-term trends in ageing related public spending, notably on pensions, health and long-term care. This will be based on the latest demographic projections by Eurostat.

PART TWO : A NNEXES

2. Originele weergave

Open document als PDF

3. Meer informatie

verder

Commission Staff Working Document - Demography Report 2008:Meeting Social Needs in an Ageing Society - Hoofdinhoud

Inhoudsopgave van deze pagina:

Delen

1.

Tekst

2.

Originele weergave

3.

Meer informatie

Zoeken:

Laatste nieuws:

publicatiedatum	26-11-2008
kenmerk	16320/08