Changing family life in Europe: Significance for state and society Open Access

Europeans have always held high expectations about the kind of family lives that they wish for themselves. Moreover, they seem intent on realising their expectations even if these differ from the model of family life envisaged by the state. This article investigates how the forms and meaning of family are changing in Europe, in their own right and in the context of the prime orientations of family policy. My intention is, on the one hand, to identify the main changes characterising family life and the risks that arise from these changes and, on the other, to problematise how the state is responding to these risks. The article is intended to provoke thought on the matter of the future of the continuously evolving relationship between family, state and society. It should be read, therefore, along the lines of an overview of the key issues on the family landscape in Europe and the dilemmas or challenges that they raise. The article is divided into three sections. The first describes the main changes and trends affecting family life in Europe in the last decade or so. The second seeks to identify the crucial risks that arise from these changes and the third draws out some resulting challenges for the state and public policy.

Before beginning the discussion of change, we should take note of a primary insight from existing work which cautions against viewing family as a receptacle, in the sense of a passive object of changes occurring elsewhere. To quote Strohmeier (2002: 344): ‘families are remarkably autonomous, self-determining social systems’. Scholarship today highlights how family is itself a wellspring of change, especially in regard to family-related roles and relations among family members (e.g., changes in parent child relations, changes in spousal roles). With this in mind, this article advances from the premise that a differentiated approach to family change is needed. In contemporary discourse when reference is made to family change it is usually the structure or form of families that is alluded to. I suggest that this is too narrow and so adopt a three-fold framework to analyse family change. This differentiates between change in the structure or form of families, in family organisation, and in family relationships and values. In essence then, demographic changes are seen as associated with and accompanied by changes in social practices, social relations and values. It should also be noted that since the comparative canvas is very large – the EU member states – the analysis undertaken is by necessity broad brush in nature. The review that follows makes no claim to completeness but is sufficient to point to significant aspects of family change in Europe at the present time.

As economic welfare increases in more countries, people have less need to share their living arrangements and be part of the same household (De Jong Gierveld 1998: 31). One very strong trend in Europe overall is towards living alone. The number of household units in Europe has increased sharply and is predicted to continue on an upward curve. In 1961 the EU-15 had 92 million households with an average of 3.3 persons per household; by 1995 the figure had risen to 148 million with the average per household at 2.5 persons (EUROSTAT 2003b). The main cause of the increase is the sharp rise in the number of persons living alone – there are now 42 million persons in this living situation across the EU-15 countries, representing some 28 per cent of the total population. Solo living is primarily a Northern European phenomenon. Although the number of one-person households has grown in almost linear fashion since the 1960s in practically all EU member states, Germany and Finland have seen the largest increase. Looking to the future, Ireland, closely followed by Spain and Luxembourg, is predicted to record the largest increase in one-person households in the coming years (ibid). This scale of living alone is in Europe historically unique. The trend is associated with a fundamental change in both the structure of households and the life course. Key in this respect is a move away from multi-generational households and an intensification of a trend towards individualisation of the life course.

A second notable trend is a fall in fertility. While varying in intensity from region to region, this is a robust pattern and has made under-replacement fertility levels the norm in Europe. Between 1980 and 2003 the total fertility rate in the EU-25 fell from 1.88 to 1.48 (EUROSTAT 2004b). As is well known, the countries of Southern Europe (in particular Greece, Italy and Spain) have not only seen a large fall in the shortest period but they have the lowest replacement rates in the European area. They are now joined by the new EU member countries in Central and Eastern Europe. This part of Europe had remarkably high and generally stable fertility levels until the mid-1980s, after which began a decline that has accelerated steadily over time. Today Slovakia and the Czech Republic have the lowest replacements levels in Europe. The change in this part of Europe has not only been very rapid but has occurred over a shorter period than in Western Europe. Ireland occupies the opposite end of the continuum – at 1.98 Irish fertility is considerably in excess of that of both the next highest country (France at 1.89) and the EU average (1.48). While falling fertility tends to be equated in the popular mind with increasing childlessness, delayed family formation and decreasing propensity to marry are also involved. Indeed, the increase in the mean age of childbearing that results from the ongoing postponement of births (the so-called ‘tempo effect’) leads to a significant and lasting loss of births that greatly contributes to population decline and ageing in Europe (Lutz 2004). Falling fertility too feeds into the move towards smaller households. It has also, together with an increase in life expectancy, resulted in a dejuvenation and aging of the population (De Jong Gierveld 1998: 38). This has meant in some countries a shift in the age distribution.

A third trend that points away from the past is a downward movement in the number of marriages. Kaufmann (2002: 423) speaks of a growing disinclination towards marriage and an increasing social recognition of forms of partnership and parenthood outside of wedlock. For the EU-25 the crude marriage rate fell from 6.7 to 4.8 between 1980 and 2003 (EUROSTAT 2004b). It is hard to discern a regional pattern here, however – Denmark, for example, has not only seen an increase in the marriage rate since 1980 but as of 2003 had the highest crude marriage rate in the EU, whereas neighbouring Sweden experienced a small fall and had the lowest marriage rate of the EU-15. The overall trend in divorces, which is upwards, contributes also to the destabilisation of marriage (as well as to a process of individualisation of childhood). The crude divorce rate per 1,000 population increased from 1.5 in 1980 to 2.0 in 2001 (ibid). Matthijs and Van den Troost (1998) characterise what is happening as a ‘divorce explosion’ and underline the fact that the high rate of divorce that now exists in many countries is completely new. Unlike marriage, there is a discernible regional variation in the prevalence of divorce and it is in the expected direction: divorce rates are lowest in the Southern countries and highest in the Czech Republic, the Baltic states and Scandinavia. Alongside and associated with these changes, there has been a growth in cohabitation. The sociological distinctiveness of marriage and cohabitation should be underlined. Marriage is typically associated with long duration, a high level of sharing and exchange of resources and heavy institutionalisation. In considerable contrast, cohabitation is open as regards duration, conditional about the amount of sharing involved and has more the character of a private arrangement than a public institution (although there are obvious moves underway to increase rights for cohabitees).¹

All of these changes are leading to and are accompanied by increasing variation in the composition of households and families. This trend has different constituents. One contributory factor is an increase in extra-marital births. In 2003 30.2 per cent of all births in the EU-25 were outside of marriage (compared to around 9 per cent in 1980) (EUROSTAT 2004b). This spells different arrangements within and across countries. One trend that is quite widespread is the growth of lone parenthood. For example, in 2001 lone parents comprised 9 per cent of all households with dependent children in the EU-15 (EUROSTAT 2004a). The underlying process of change that is involved is encapsulated in the literature by the concepts ‘new family forms’ and ‘new biographical models’. These are meant to refer, on the one hand, to the growth of cohabitation or partnership (including among same-sex couples) and the many combinations of family type that these and other changes engender and, on the other, to a change in people's life forms and biographical sequencing. In effect, the interweaving of marriage, sexuality and procreation is being unpicked (Matthijs and Van den Troost 1998: 112).

Looked at overall, while there are strong similarities, it would be ill-advised to make a simplified argument of European convergence, not least because the starting points are different and convergence implies similarity of outcome. Boh's (1989) term ‘convergence to diversity’ continues to have resonance. In terms of trends, countries are moving at their own pace and developments are embedded in national cultures and traditions. Kuijsten (2002: 50) summarises well with his observation: ‘pluralization is everywhere but everywhere it has another face’. In terms of patterning, cross-country differences persist. One must, at the minimum, draw attention to the fact that Europe has (at least) two kinds of pattern in relation to family composition and structure. These tend to follow (loosely) a North/South continuum. In Northern European countries, especially Scandinavia, in comparison to those further south more people live alone, average household size is smaller, marriage is less common and alternative living arrangements more diverse and prevalent. Whether there is also an East/West gradient of change is something that needs to be investigated as developments unfold – at the moment patterns in Central and Eastern Europe appear to be in flux and so it is difficult to characterise them definitively.

A move to two-income families captures a primary line of change in this regard. One of the main motors of change is the increase in labour market participation of women. This, although not completely independent of the state of the economy, has in fact been one of the most dominant and persistent trends to be seen in Western European countries over recent decades (Rubery et al. 1998; Daly 2000; EUROSTAT 2002a). The movement in participation levels is inexorably upwards. While a prediction of the disappearance of the housewife might be premature, increased employment among women is associated with a move towards two-income families and a decline in the male breadwinner/female homemaker household arrangement. The two-income family is now the dominant form of household in most EU-15 member states among households with two people of working age. For the 10 member states for which comparable data are available, households with both partners in the labour force were in 2000 almost twice as numerous as those with only one, averaging around 62 per cent of the total (EUROSTAT 2002a). A marked divide is evident, however, between the northern member states, together with Portugal, where two-thirds or more of households were dual-participant, and Spain, Greece, Ireland and Italy where the proportion was less than 50 per cent. In both groups of countries, however, the 1990s saw an increase in the prevalence of dual-participant households. The growth was particularly pronounced in Ireland, The Netherlands, Belgium and Spain. Dual participation is increasing most among couples with children. However, as always, one must probe the amount of participation, in terms of the hours worked. The available data reveal considerable variation, although the most common form of dual participation in all countries apart from The Netherlands² is one where both participants work full time (ibid: 3). The one-and-a-half model – where he works full-time and she part-time – is found in about 30 per cent of all couple households in the 10 countries.

This is a far-reaching change in a number of respects (Lewis 2003). It signifies first a process of alteration in the relationship between family and economy, and in particular between family and employment. In addition, the relationship between family and society is being recast since the breadwinner/homemaker model is far more than an economic arrangement, spelling also a particular form of relationship between spouses/partners, an arrangement for the care of children, a pattern of intergenerational relationships and a division of labour between state and family. Continuity sits alongside change though, for part and parcel of these trends – and a feature of the evolving two-income family form – is a (continuing) inequality in the distribution of household work. In other words, there is little evidence of the emergence of household working-time arrangements that are compatible with a more equal sharing of paid and unpaid work. The disparity is mirrored especially in the distribution of unpaid work among women and men. However this is not a pattern that is set in stone. There are certainly some social class variations involved: the educational attainment of women has a significant effect on whether they as well as their partners are employed and whether they have part-time or full-time jobs (EUROSTAT 2002a: 5). Women with higher education are more likely to be members of dual-earner households as compared with their less well-qualified counterparts.

Not alone has the numerical composition of households fallen sharply over the past century but the lifestyles and relations of those who share the same household have also changed. Research suggests that there are fundamental shifts underway in the social organisation of intimacy and sociability and that the trend is, on the one hand, towards increasing individualisation and, on the other, towards increasing diversity of relationship practices (Giddens 1992; Beck and Beck-Gernsheim 1995). The emotional interior of the family has changed as a result. In comparison to the past, families today are arguably more complex, more heterogeneous and more often in flux and mutation (Matthijs and Van den Troost 1998: 111). A key aspect of change is that the balance of power within families is altering. ‘The modern family in most European countries has turned from an authority-oriented family to a negotiating one’ (Du Bois-Reymond 1998: 59). This has at least two dimensions, affecting relations between partners/genders and those between parents and children. Women have gained greater power in their relations with men, a development characterising intra-familial relationships as well as those outside the family. While it is premature to pronounce a radical change in this regard, there are distinct signs of a move towards more egalitarian familial relations. Both women and men are now normatively oriented to a combination of family roles and participation for both in paid work, although the degree of sharing of household work has not radically altered. Another aspect of family life that is changing is the role of children. Occasioned by general social change and expedited by the tendency for states to grant children autonomous rights, children are increasingly seen as and enabled to be agents in their own right (Brannen 1999). It is important not to overstate the degree of change that has taken place in this regard either, however because patriarchal norms have a long shelf life.

One of the most significant developments is the emerging divergence between partnership (or coupledom) and parenthood. A growing divide is to be observed at the level of desires and emotions as well as at a more practical level in terms of life organisation and trajectory. It seems that partnership and parenthood are not just two different types of social relationship/institution but diverging life choices. As evidence, consider the growth in the proportion of partnerships or marriages without children and how being in a partnership is itself now a legitimate ‘family end’. In sum, we see a move away from the biographical pattern of love relationship leading to marriage and then, more or less immediately, to child bearing, to one characterised by a sequence of romantic relationship and partnership/cohabitation to parenthood that may include marriage or to continued partnership without children (Tyrell and Schulze 2000). These authors suggest, further, that partnership and parenthood are becoming increasingly incompatible. Partnership demands mobility and typically is not founded on a long-term commitment, whereas the increasingly child-centred family of today requires immobility and stability from parents who are put under growing pressure around the quality of their parenting.

There is no doubt but that families present a complex environment for policy because, as Gonzalez-Lopez (2002: 23) points out, the map of living arrangements and the individual life cycle has become more difficult to predict and read. However while a wider range of options for private living has opened up, it is also true that the current changes are patterned (lending some assistance to the analyst and policy maker). Cantillon (1998) is correct to point out that a new relationship is being forged between needs, risk and risk coverage. This is true in two senses: the occurrence of the risk events traditionally covered by the social security system (unemployment, sickness, old age, death of a breadwinner) no longer automatically triggers a situation of need and new risks have arisen which are not covered by social security and income policy. Social protection, she says, has to be adapted to the new family context (1998: 230). What risks are we talking about?

Taking patterns together, and focusing on their implications for state, family and society, I suggest that there are four key family-related ‘risks’ in contemporary Europe.

If in the past policy could take the existence of the family for granted, it can no longer do so. For people's readiness to form families at all is now at stake. One result is that Europe has a shrinking family sector and that children have become a relatively scarce resource. It seems that couples today weigh up the pros and cons of having children through a process in which a complex calculus (a mix of material and self-development/self realisation considerations) looms large. To understand the process, it is insightful to take the perspective of the young couple. Huinink (1997), cited in Kaufmann (2002: 451), identifies three problems that young couples face before they commit to parenthood: the problem of co-ordinating the long-term perspectives of both partners; the problem of insufficient resources to have children; the problem of the compatibility of public and private commitments, especially in regard to the relationship between work and family. In sum, children are fitted into life plans of adults and the ‘having’ of children is potentially in conflict with the achievement of other goals (Jensen 2003).

Looking more closely, attention turns on the extent to which people feel that their circumstances are in line with their desires. There is evidence that people across Europe wish to have a larger number of children than they succeed in having. A recent study reports a gap between people's actual and ideal family size (of about 0.29 children for the EU-25) (Alber and Fahey 2004: 45–6).³ Circumstances are perceived to be preventing people from having more children. When women who do not fulfil their fertility aspirations were asked why they had fewer children than they wished, they pointed mainly to reasons of a broad economic character. The authors are keen to point out that this is less a matter of a lack of resources in an absolute sense than of opportunity costs in terms of a woman's time and career in the labour market. This kind of interpretation is supported by other work, especially research that seeks to link fertility to policy provision. This is a notoriously difficult relationship and so any suggested patterns need to be treated with great caution. However, one can make a general case that countries with higher gender equality exhibit higher fertility scores than those where women find it difficult to reconcile an independent life with family obligations (Kaufmann 2002: 450).⁴ Moreover, a positive relationship is also observed between the supply of child care and fertility. Hence, it appears that a somewhat counterintuitive relationship prevails: the more ‘modern’ is the provision the higher the likelihood of having children.

The desired family arrangement and the chances of attaining it form another part of the explanation for falling fertility. The evidence suggests that people in Europe do not have what they want in this regard. Despite significant focus on reconciling work and family life and policy reform towards this end, there exists a wide divergence between the actual employment/family arrangements that people have and those that they would prefer (OECD 2001). In general, across Europe the model that people have too much of is the traditional male breadwinner model (of employed father and home-making mother). In Germany, for example, this is the actual arrangement of 52 per cent of the population but for only 6 per cent is it their preferred arrangement; similar levels of dissatisfaction with employment/family arrangements are to be found in Italy. In general, the two-earner family form is more sought after than the traditional model of a male breadwinner/female caregiver. The model that is considered as too seldom available across Europe as a whole is the ‘one and a half earner’ arrangement whereby the man works full-time and the woman part-time. One has to ask how sustainable such a gap between expectations and reality is, especially in countries such as Germany and Italy, and to a lesser extent France and Ireland, where the opportunity for people to realise their preferred family/employment arrangement seems to be quite compromised.

In many countries there is occurring a dislocation between partnership and family. This is not equivalent to an ‘us and them’ development – partners versus families with children – but it seems as if they are two quite different sectors of the population and two different styles of life. Change is driven by the younger age cohorts and the upper social classes – the likelihood of wanting or having children is therefore structured around the two axes of age and social class. To risk some exaggeration, it seems that Europe's younger generations, especially those from the upper classes, are prepared to forego family life with children. In the event, the task of reproduction is left increasingly to the lower income groups, and to immigrants. There is a polarisation taking place and it is multi-layered. For example, in regard to gender, women in almost all countries face large (although cross-nationally varying) trade-offs in combining employment and motherhood. In regard to socio-economic differences, it is now the case that families with children have a much higher likelihood of being one income as against two income. In regard to nationality, there is a widening gap between fertility rates, family size and income levels of immigrants vis-à-vis nationals. So what one has is a layering of the presence of children and family size around age, income and education and ethnic origin.

Whether one or both partners works outside the home has also become a (new) factor in social inequality. As Matthijs and Van den Troost (1998): 113–4) put it: ‘some of the single earners are elbowed into lower income groups while the double earners climb into the higher prosperity groups’. Double income, therefore, has become the wealth norm today. The main reason why couple families with children are one income is because the second partner is involved in care. There is no inevitability about this however – at least some families are low income because care, or unpaid work more generally, is not considered either a legitimate social risk for social security and other policy purposes or as gainful work in its own right.

The ultimate outcome may be a process which Strohmeier (1993, cited in Schulze and Tyrell 2002: 88) describes for Germany as a movement towards class-specific family structures. What we are seeing, then, is an interweaving of social inequality with family whereby the upper income groups are increasingly less children-oriented and the presence of children is ever more closely associated with social class position.

There are different ways to state the contemporary gender problematic. A relatively benign characterisation is that the modernisation of gender roles and relations is incomplete, whereas a more trenchant interpretation is that women are so disaffected by the current scenario that they are effectively on a baby strike. One of the main problems is that the movement of women into the labour force has not resulted in a sharing of home-based or labour-market work between women and men. Women continue to be responsible for the bulk of work in the home; the average woman doing between two and three times the amount of unpaid work carried out by men (Gershuny 2000). In addition there is a large disparity in the hours worked by employed women vis-à-vis their male counterparts. ‘Work/life balance’, such a widely-used term today, takes on a completely different meaning in this context, suggesting that the task of ‘reconciliation’ is a female responsibility.

Recent research in Ireland reveals the kind of complexity involved. As well as logistical difficulties in making arrangements and finding care for their children, women are finding it increasingly difficult to combine their two worlds (Daly 2004a). A national consultation exercise with more than 700 people found that the role of mother is far from settled and is a source of considerable tension (if not dissatisfaction) on the part of women. Emotional difficulties are rife. Ambivalence is the lot of many mothers and they feel torn between children and work. Against a background of massive increase in female labour force participation in the last decade – resulting from buoyant economic growth and state policies that encourage women into employment – one of the most strongly and consistently expressed views was that mothers must have more options or choice around whether they want to take up paid employment or not and the conditions under which they make the choice.

As is well known the dejuvenation of the population and the stretching of life at the upper end raises issues (for states, families and individuals) around care. There is a series of risks involved. One such risk is that states will not be able to (afford to) provide quality care in the volume needed, mainly because of insufficient resources. A second is that family members will not be able to care for or see their relatives cared for in a manner that satisfies them. In this regard it is important to note that a strong ethic of informal and family care is integral to the European value frame and that informal care is quite widespread. Recent research by the European Foundation for the Improvement of Living and Working Conditions (Alber and Fahey 2004; Alber and Köhler 2004) shows that about four out of five people across the EU-25 consider it a good thing to strengthen family responsibility in looking after elderly persons. Domestic care for older people is almost ten times more popular than residential care. This same research reveals a remarkably vital network of informal help throughout Europe. In acceding and candidate countries, roughly a quarter of respondents are engaged in some form of regular help of others; in the EU-15 the proportion is about 21 per cent.⁵ These results lead Alber and Köhler (2004: 70) to speak of a ‘strong and rather unbroken tradition of family support in Europe’. Moreover, the enlargement of the EU strengthens the ethic of family care in Europe.

Among the issues that states have to keep a close eye on is the production of welfare, and especially care, within the family. In this regard, the ‘care potential’ is critical: the availability and ability of people to care for their relatives. Research on care patterns suggests that what really matters in this regard is the presence of spouses (de facto the main carers of elderly people) and of daughters and sisters (De Jong Gierveld 1998). Decreasing population and increased mobility reduces the care potential. Willingness to care of course also comes into question and here what seems to matter (in terms of the next generation's willingness to care) is the quality of the relationship between parents and children. Overall however, the most recent research indicates that intergenerational solidarity is strong in Europe, with the younger generation expressing a wish to undertake care for their elderly relatives, just as the current older generation seem willing to shoulder their part of the costs of care and do not advocate shifting the cost to the younger generation any more frequently than do the young themselves (Alber and Fahey 2004).

The role of the state is central in all of this. One can characterise the key challenge of and for family policy as being to reconcile the modernisation of family relations with the economic, social and demographic needs of the society (Kaufmann 2002: 462). However, it is an open and difficult question of how much the state should ease the passage of families. Arguments against an interventionist approach are easily marshalled. If families are left to their own devices they might be better able to weather whatever storms they encounter. Interventionist policies may also be charged with social engineering. However, counter arguments are also compelling, not least the claim that the development of a family policy is integral to the future of European market societies. In addition, the fact that people have high expectations about their family lives – and that many consider these expectations to be unfulfilled at present – may render it impossible for governments to avoid instituting policies to effect particular outcomes in family life (Alber and Fahey 2004).

In any case, states have not been indifferent. While it may not always be named as ‘family policy’, the domain of family is an expanding area of policy intervention in Western Europe and the ‘family question’ is to the fore in contemporary public debate.⁶ Indeed, as Martin (2004: 14) points out, the ‘family question’ is coming to be seen as a component of the ‘social question’ whereby the family's contribution to social order, social stability and even social inclusion is increasingly debated. In an earlier article (Daly 2004b), I identified a number of characteristics of recent policy in Western European countries as they target and respond to family change. While we need to underline that there is no single or uniform response on the part of states and indeed that historically Europe was characterised by different types of family policy, it is possible to discern some general trends in current policy.

As I read it there are two lexicons currently informing policy in Europe as it relates to family. One is an instrumentalist type of discourse whereby the primary orientation of policy is to address particular ‘problems’ of family life (mothers’ (under)employment, paternal responsibility, the quality of parent–child relations, family poverty, child well-being, family stability, social order). A second lexicon – dominated by a liberal language of ‘quality’, ‘choice’ and ‘self-fullfilment’ – now sits alongside the first. This is informed especially by the recognition of disaffection with some of the conditions of child-rearing in today's market society (low income, lack of opportunity for women to be employed and at the same time reduced opportunity to care for family members). It has another side too, though, in that it both emerges from and lends itself to a critique of how people, parents especially, manage their family lives. This kind of discourse serves to justify a focus for social policy on the ‘performance’ of family roles and responsibilities and greater regulation on the part of the state of the nature and quality of family-related activities and relationships.

In terms of concrete policies, the big story now is of an ever closer relationship being forged between family policy and employment policy. Measures to achieve the ‘reconciliation of work and family’ are now the dominant theme threaded throughout European and especially EU policy on the family–work relationship. The most widespread framing of this as a policy objective is, on the one hand, to enable workers to attend to some of their family-related concerns or responsibilities and, on the other, to ensure that family exigencies or desires do not get in the way of able-bodied adults becoming and remaining paid workers. Tax credits, employment leaves for parental and care purposes and growing childcare provision are the flagship policies oriented to this end. Eschewing the expansion of general or generic subsidies to families with children, financial support to families nowadays is directed towards specific employment-related costs such as those associated with service procurement or the loss of (potential) wages. This is a significant change because traditionally in Europe cash support to families was oriented to assisting them with the direct costs involved in rearing children (those associated with food, clothing and education for example).

A related trend in recent family policy is for one type of ‘model family’ to be replaced by another. One could put this otherwise: the problem of the false generic in family policy continues. Historically the most widespread family model underlying social policy in European countries was the breadwinner/homemaker model (whereby the man worked outside the home and the woman within it). The cardboard cut-out family of policy today is different but equally particular: it is the working family, wherein men fulfil their obligations to be active fathers emotionally and economically and women act as good citizens by being employed. The imprint of this kind of family model is to be seen in policies that encourage and normalise employment for both parents.

Thirdly, there is a movement away from some of the old principles underlying family policy. Traditionally, family policy was quite a distinct domain of social policy in Europe, with emphasis (varying across national settings) on promoting fertility, addressing poverty and seeking horizontal equity (the latter in the sense of compensating families for having children by reducing income disparities between them and families without children) as well as, more recently, gender equality (Wennemo 1994; Gauthier 1996). Over the course of time, pro-natalism has waned as a policy orientation, just as equity and anti-poverty are generally less visible as principles of family support in Europe today as compared with the past (although they appear quite prominently in the rhetoric). In particular, the emerging policy consensus around ‘reconciliation’ has acted to blur concerns about equity and, even in some cases, to corrupt the traditional compensation and assistance function of family policy (it is not uncommon nowadays to hear child benefits being referred to as a payment for the care of children for example). Gender equality is also less emphasised in contemporary family policy (although it must be said that, outside of Scandinavia, it was never as deeply entrenched in family policy as some of the other principles). This is not to say that family policy and social policy are not concerned with matters relating to gender – they are – consider the emphasis on getting fathers more involved in the life of their children through the expansion of paternity leave and also parental leaves. However, the language and problematic of contemporary social policy tend towards gender neutrality and so gender differences or inequalities or the specific situation of women are not per se being problematised for the purposes of social or family policy (Daly 2004b).

The material presented throughout this article underlines the need for diversity in policy response. This is not what is happening. For all the diversity that there is on the family/household front and in the types of approach taken to the family by social policy historically in European welfare states (Saraceno 1997), policy on the family today is, I would argue, becoming narrower in focus. Witness the focus on a singular family model, the closer relationship between family policy and employment policy, the narrowing as compared with the past of some of the key principles of ‘family policy’. Overall, and not for the first time, a series of disjunctures are to be observed between family life and family policy. Europeans are being let down in three key ways by current policy.

The first is in regard to the financial security and well-being of families with children. As we have seen having children is associated with income constraint and relative disadvantage vis-à-vis other sectors of the population. Considered as an issue for policy, the critical challenge here is redistribution and to achieve a balance between horizontal and vertical equity. Given the ever closer relationship between large family size and low income, there is a need for policies to focus more closely on vertical equity. In addition, as Alber and Fahey (2004) and others point out, measures must also have a horizontal cast (in terms of redistributing resources and opportunities between those with children and those without). An anti-poverty orientation would also appear to be essential, given that there is a growing generational gap in poverty, to the detriment of children (Clarke and Joshi 2003).

A second issue is around fertility and child-rearing (which should be seen together). These are by no means easy issues for policy, although looked at historically Gauthier (1996) suggests that concern for declining fertility has been the strongest motive for governments to take action on matters of family policy. The evidence available suggests that pronatalist policies would have greater effect if targeted (also) on those for whom the opportunity costs of children are highest (women with high earning power). The likely effectiveness of an approach that focuses solely on more generous cash benefits is low in that allowances can never be high enough to motivate someone in the higher income category to have a child. As Strohmeier (2002: 355) says in relation to Germany: ‘Despite the particular emphasis on economic intervention in the national policy profile, lack of money is a bottleneck only for those who already have children, but it is not the main reason to be childless of those without children’. The main problems are, rather, of restricted social and economic participation of women, the continuation of relative inequality in the internal division of labour in families and constrained opportunities for people (men as well as women) to care for their families.

The third failure is an inability, or unwillingness, on the part of policy to feed people's appetite for quality in their family lives. Quality is denied people most often because of stark trade-offs, especially between family and labour market or children and partnership. However it must be pointed out that what people understand by quality is not by any means unidimensional – it is closely related to diversity in the sense that what constitutes quality may change as family life progresses. Policies which ease trade-offs are necessary as are those which recognise a diversity of family life situations across the life course. It seems at the moment that the state is stuck in a groove of defamilialisation (of women especially) and marketisation. These, the main ideas of recent years, are unlikely to prove a sufficient response. At root is a profound set of questions about the assumptions that state actors make about family and the relationship between family, market and state. They do not appear now to be any more capable than they were in the past of delivering on people's preferences in relation to family life.

Mary Daly is a Professor of Sociology at the School of Sociology and Social Policy at Queen's University Belfast. Among the fields on which she has published are poverty, welfare state, gender, family and the labour market. Much of her work is comparative, in European and international context. Among her recent relevant publications are Care Work: The Quest for Security (ILO, 2001), Contemporary Family Policy (with S. Clavero, IPA, 2002), Gender and the Welfare State: Care, Work and Welfare in Europe and the USA, (with K. Rake, Polity Press, 2003)