Journal of European Social Policy http://esp.sagepub.com Patterns of paid and unpaid work in Western Europe: gender, commodification, preferences and the implications for policy Jane Lewis, Mary Campbell and Carmen Huerta Journal of European Social Policy 2008; 18; 21 DOI: 10.1177/0958928707084450 The online version of this article can be found at: http://esp.sagepub.eom/cgi/content/abstract/18/1/21 Published by: ($)SAGE Publications http://www.sagepublications.com Additional services and information for Journal of European Social Policy can be found at: Email Alerts: http://esp.sagepub.com/cgi/alerts Subscriptions: http://esp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 27 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://esp.sagepub.eom/cgi/content/refs/18/1/21 Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Article Patterns of paid and unpaid work in Western Europe: gender, commodification, preferences and the implications for policy Jane Lewis*, Mary Campbell and Carmen Huerta, London School of Economics, UK Summary This article explores how parents in couple families reconcile employment and child-care, and how far the current emphasis of EU-level policy on enhancing the formal provision of child-care fits with patterns and/or preferences in Western European member states. We use European Social Survey data from 2004-05 on working patterns and preferences, and on child-care use and preferences regarding the amount of formal provision. We find that working hours remain a very important dimension of work/family reconciliation practices, with large differences in both patterns and preferences. There is very little evidence of convergence towards a dual, full-time worker model family outside the Nordic countries, although the balance between the hours which men and women spend in paid work is becoming less unequal. The part that kin (partners and grandparents) play in providing child-care remains important in all but three countries, and, for the most part, mothers report that they are content with the amount of formal child-care available. We suggest that work/family reconciliation measures need to encompass a more extended policy package, the components of which are likely to be specific to member states. Key words child-care, EU-Lisbon strategy, European Social Survey, mothers' employment, work/ family reconciliation women tended to polarize between working full-time or not working at all. Scandinavia is now close to a full-time, dual-earner family model, supported by the state via paid parental leaves and extensive formal care services, though with women still working shorter hours than their male partners. Changes in male and female contributions to families have resulted in 'new social needs and demands, labelled 'new social risks' (Bonoli, 2005: 431): notably, reconciling or combining work and care. This is now a challenge common to all European welfare states. Since the late 1990s, a prime aim of EU policy has been to get more women into employment. In 2000, *Author to whom correspondence should be addressed: Jane Lewis, Professor of Social Policy, London School of Economics, Houghton Street, London WC2A 2AE, UK. [email: j.lewis@lse.ac.uk] Journal of European Social Policy 0958-9287; Vol 18(1): 21-37; 084450 Copyright © SAGE Publications 2008, Los Angeles, London, New Delhi and Singapore, DOI: 10.1177/0958928707084450 http://esp.sagepub.com Introduction By 2000, the 'traditional' male breadwinner model family had been substantially eroded throughout the European Union, although to different extents. In north-western Europe, the increase in women's employment - not matched by an increase in men's care work and housework - had resulted in the emergence of a 'one-and-a-half or 'one-and-three-quarters' earner model. Men worked full-time and women worked varying numbers of part-time hours, with a large amount of care work remaining informal (Hobson, 2004; Lewis 2001). In Southern Europe, part-time work was rare and Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 22 Lewis et al. the Lisbon Council set a target of 60 percent female employment by 2010, which was supplemented the following year by an interim target of 57 percent by 2005. Significantly, neither target specifies hours for women's paid work. EU policy also shifted perceptibly towards encouraging child-care services rather than leaves and in 2002 the Barcelona Council set targets for child-care provision to reach 90 percent for children between three years old and school age and 33 percent for under-threes. However, these policy developments appeared to be instrumental: they seemed to serve first and foremost the agendas of competition, growth and the budgetary implications of the worsening dependency ratio, rather than family welfare, child well-being, parental choice or gender equality per se (e.g. Lewis, 2002). This instrumentalism seemed to intensify after the mid-term report on the Lisbon Strategy (Kok, 2004). As a recent Commission consultation document on extending EU legislation on leaves, care services and working time states: 'In March 2006 the European Council stressed the need for a better work-private life balance in order to achieve economic growth, prosperity and competitiveness and [to this end] approved the European Pact for Gender Equality' (EC, 2006: Introduction; see also Fagan et al, 2006). Member states have also sought to increase female employment, though many have increased child-care leaves alongside child-care services and member states have at different times prioritized a much broader range of policy goals than has the EU. As academic commentators from various disciplines have pointed out, policies must provide for paid leaves, care services and 'family-friendly' working hours if reconciling work and care is to be family-centred and promote gender equality (Bettio and Plantenga, 2004; Gornick and Meyers, 2003; Leitner and Wroblewski, 2006; Lewis, 2006). However, family practices and policies are particularly influenced by norms and values (Strohmeier, 2002), and respect for choice in the intimate arena of family relationships is a sensitive area for policymakers wishing to transfer work from the unpaid to the paid labour market - to commodify labour - through more female employment and formal child-care provision. The latest European Social Survey (ESS: 2004-05 data) provides information on working patterns and working hours preferences and, for the first time, on the use of formal and informal child-care and how far respondents would like more formal provision. Journal of European Social Policy 2008 18 (1) (It also provides information on the use of child-care leaves, but not on leave preferences.) We have analysed ESS employment and child-care data for 13 EU countries in order to explore how parents reconcile employment and child-care, whether a policy focus on better formal child-care provision in particular is likely to fit with patterns and/or preferences in member states, and the extent of likely support for policies beyond increasing formal child-care provision and enforcing the maximum 48-hour week specified in the 1993 Working Time Directive (93/104/EC). Work, care and commodification: patterns, preferences and issues Wallace (2002) has argued that increased female labour market participation makes new household strategies necessary; certainly, patterns of paid and unpaid work at the household level in member states are complex. In respect of care, mothers and fathers may juggle working hours, with the help of state legislation (e.g. on the right to flexible/part-time working); one parent (usually the mother) may just 'scale back' her employment/career for a shorter or longer period and with or without financial compensation (Becker and Moen, 1999); fathers may do more at home by working atypical hours or using leave entitlements where these exist; other informal care may be used; and care may be more or less com-modified. Different patterns are accompanied by different kinds of gender divisions. It is possible, for example, for a high rate of labour participation among mothers to be facilitated to different degrees by firms, by legislation, and by men doing more in the household. Patterns of female paid work vary hugely in Western Europe and do not fit Esping-Andersen's (1990) 'welfare regime' types (Daly, 2002). Table 1 shows men's employment rates rising slightly in most countries since 1994, and more substantially (5+ percentage points) in Finland, Ireland and Spain. Women's employment has risen everywhere, with large increases of around 10 percentage points in Belgium, Finland, Greece, Italy, the Netherlands and double that in Ireland and Spain. Eight of our 13 countries had achieved the EU's Lisbon target of 60 percent female employment by 2005. Part-time employment (defined in Table 1 as under 30 hours a week) is very important in the Netherlands (above Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Table 1 Employment rates for men and women and the gender employment gap, 1994, 2000 and 2005a (% of men/women aged 15-64 years) Men's Employment employment Women's employment gender gap ALL ALL Part-time FTE FTE (a) 1994 (b) 2005 (c) 1994 (d)2005 (e) 1994 (f) 2005 (g) 2000 (h) 2005 (j) 2000 (k) 2005 Austria 78 75 59 62 nab 29.6 51 50 25.2 22.0 Belgium 67 68 45 54 30.0 33.1 44 46 26.5 21.9 Denmark 78 80 67 71 26.2 24.9 62 61 14.7 14.5 Finland 61 69 59 67 11.5 14.8 61 62 8.8 6.4 France 66 68 51 57 24.5 23.3 49 51 20.4 16.1 Germany 74 71 55 60 28.0 39.4 46 45 25.0 23.3 Greece 72 75 37 46 13.1 11.1 41 45 31.4 30.3 Ireland 65 76 39 58 25.5 34.8 45 49 31.0 27.1 Italy 68 70 35 45 20.6 29.2 37 40 30.3 28.7 Netherlands 75 77 53 65 54.5 60.9 41 42 34.2 30.0 Portugal 74 73 55 62 15.2 14.4 57 58 19.2 15.1 Spain 63 76 32 52 14.3 22.2 38 45 32.9 28.6 Sweden 72 76 71 72 24.9 19.0 60 61 9.8 10.6 UK 75 79 62 67 41.2 39.3 50 52 24.5 21.8 EU-15 71 73 49 58 28.3 32.3 45 47 25.7 23.0 Notes: a Because almost all countries have revised their series since 1994, comparisons over time and between countries are indicative rather than precise. In this table, 'part-time' employment refers to persons who usually work less than 30 hours per week in their main job. Columns (e) and (f) show the % of women who work PT as % of all employed and self-employed women in the OECD sample who declare their usual hours of work. FTE = Full-time equivalent: i.e. total hours worked divided by the average annual number of hours worked in full-time jobs, calculated as a proportion of total female population in the 15-64 age group. The FTE employment gap is the difference in employment rates measured in full-time equivalent between men and women in percentage points. b Using a different definition, part-time employment of women in Austria is reported to have risen from 28% of total female employment in 1997 to 40% in 2004. Accessed on 16.1.07 at [www.eiro.eurofound.ie/2006/10/articles/at06100491.html]. Sources: Columns (a)-(f): Statistical Annex of OECD's Employment Outlook 2006 and 2007 (Tables B and E). Columns (g)-(k): European Commission's Indicators for Monitoring the Employment Guidelines, 2006 Compendium (Tables 17.3 and 18.3). 24 Lewis et al. all), but also in Germany, the UK, Ireland and Belgium, where over a third of employed women work part-time, with Austria and Italy close behind. The proportion of employed women working part-time has increased in eight countries and substantially so (8+ percentage points) in Austria, Germany, Ireland, Italy and Spain. Elsewhere it has fallen, notably so in Sweden, and perceptibly in Denmark, Greece and the UK. The full-time equivalent (FTE) employment rates in Columns (g) and (h) of Table 1 indicate the relative importance of women's employment: for example, the figure for the Netherlands, with a high employment rate and a high rate of part-time work, is lower than that for Greece, which has the second lowest female employment rate.1 In addition, FTEs can show the extent of the gender employment gap, which is bigger when women's employment rates are lower (or men's rates higher) and/or women's part-time employment is higher. Thus, the trajectory for women's employment is generally upwards (there have been hiccups - e.g. it fell sharply in Finland during the economic recession of the early 1990s). Nevertheless, patterns are hard to predict. In the Nordic countries, which have had high levels of female employment for a long time, more women now work full-time, but the Finnish recovery seems to have been partly based on part-time jobs. Among the countries with historically high rates of women's part-time work, the Netherlands and Germany - but not the UK - are continuing to increase this form of employment. The pattern of development in Southern Europe is also difficult to interpret. The expansion of female employment in Spain and Italy is based significantly on part-time work, but this is not the case in Greece and Portugal. Although national patterns vary, Table 2 shows that the presence of children continues to be an important factor associated with women's exit from employment (except in Denmark). Among our 13 countries, the employment rate of mothers is 70 percent or over in the Nordic countries, Austria, the Netherlands and Portugal, but exceptionally low in the rest of Southern Europe. In most countries, the steepest decline in mothers' employment rates occurs with the third child and having only one child under 12 has little impact. However, in Germany, the Netherlands, Spain and the UK the presence of even one child has a marked effect on employment rates. By contrast, fathers are more likely to be employed than other men in all EU countries (Eurostat, 2005). Table 2 Women's employment rates with and without children, 2003a (% employment rate for women aged 20^19, without and with a child/children under 12) PT working (% of all women) 1 child 2 children Three+ ------------------------ Without With under 12 under 12 under 12 Without With EU-25 Austria Belgium Denmark Finland France Germany Greece Netherlands Portugal Spain UK 75 83 75 77 78 77 80 57 82 77 62 83 60 72 68 80 72 66 60 53 70 76 51 62 65 78 70 80 75 73 66 54 72 78 54 68 58 66 69 82 74 64 55 53 70 75 48 61 41 57 49 67 56 40 38 40 59 60 41 38 15.2 22.7 16.8 32.3 21.8 27.2 n.a. n.a. 10.2 7.8 14.1 17.6 21.3 35.1 4.9 6.7 33.0 54.7 7.7 7.2 8.7 9.7 18.5 36.2 Noře: a Data for Ireland and Sweden not available. Full-time is defined as over 30 hours and part-time as under 30 hours. The part-time working figures refer to part-time mothers as a % of all women aged 20-49 in each country. Sources: Eurostat News Release 49/2005, Luxembourg: Eurostat; and Eurostat (2005). Journal of European Social Policy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Patterns of paid and unpaid work in Western Europe 25 In addition, while part-time work for women increases in most countries with the arrival of children, it is recognizably the way of reconciling work and family in the UK, the Netherlands, Germany and Austria. Recent Eurostat data also show that while employment rates for mothers vary with the age of their youngest child much less in the EU-15 than in the new member states, they nonetheless range from a low of about 50 percent for mothers of children aged under two in Germany, Greece, Italy, Finland, Spain and the UK to over 70 or even 80 percent in the majority of EU-15 countries when children are aged 6-11 (Eurostat, 2005). Links between policies and behaviour can be found, but are also easily undercut. Ellingsaeter and Leira's recent edited book (2006) on Scandinavia has shown the extent to which Denmark has focused support on working mothers and provides the least support for fathers; Finland supports mothers who work and mothers who care, with little for fathers; while Sweden alone provides support for working and non-working mothers and for fathers. Such policy differences seem to be reflected in family practices. Thus, in Finland, mothers of very young children are much more likely to be at home than working part-time, making use of the long home-care leave (albeit that, as Table 1 shows, on an FTE basis, women's employment is as high in Finland as it is in Denmark or Sweden). Nevertheless, major questions must be raised about any simple relationship between policies and behaviour and it is sometimes difficult to know the direction of the relationship. After all, like the US, Portugal has long achieved high rates of female full-time employment with a fraction of Scandinavia's state-supported care services, labour market factors (notably low wages and lack of availability of part-time jobs) being particularly important. The most important recent challenge to thinking about what causes patterns of paid and unpaid work among women has come from preference theory (Hakim, 2000), which suggests that personal choices, shaped above all by values, are the main determinant of behaviour. Argument has raged over how fixed preferences are, and the part played by constraint and structural factors in determining them. For example, women's level of education is a very important factor in their employment situation - highly educated mothers in the EU-25 with one or two dependent children are nearly twice as likely to be employed as are low-educated mothers (Eurostat, 2005: Table 2). Poverty, low wages and lack of access to good-quality, affordable child-care constrain choice which in turn may constrain the preferences which mothers express (e.g. Crompton, 2006; Crompton and Lyonette, 2006; Haas et al, 2006; McRae, 2003). In fact, it may well be that mothers' decisions/strategies for combining work and care depend on a compromise between the possible and the preferred (Eurofound, 2006). In all probability 'policy matters', although it is remarkably hard to prove it. Gershuny and Sullivan (2003) have demonstrated a relationship between policies and women's employment rates (see also Bettio and Plantenga, 2004), but not between policies and the domestic division of labour. Himmelweit (2005) makes a more nuanced and convincing argument that there is an iterative relationship between attitudes and behaviour, and that policies can help to change both. We are not concerned in this article with the determinants of behaviour and the nature of the relationship between preferences and practices. Rather, we use the data provided by the ESS to comment on public policy choices. It is crucial to consider the preferences that people express in relation to policy development. Visser (2002) has argued that Dutch policies to encourage women's part-time work can be characterized as 'bottom-up', in that they build on the widely perceived Dutch commitment to an 'ideal of care' (Kremer, 2006) that prioritizes parental care. Indeed, there may be resistance by women to the policy goal of commodifying care, whether because personal preferences - or at least their revealed preferences - differ (Hakim, 2000); or because of feelings about identity (Akerloff and Kranton, 2000; Himmelweit and Sigala, 2004), and 'the proper thing to do' (Finch and Mason, 1993); or because there is no experience of, or trust in, care provided outside the home. Policy in democratic states must respect their citizens' perceptions of what they want, even as it may seek to modify them. The ESS data offer information on people's practices and their own preferences regarding working hours and child-care provision. These are more robust than attitudinal data measuring people's notions of the 'ought'. There is more than one route to assisting people to combine work and family, and different patterns and value systems may indicate the need for a more nuanced approach than EU-level policy is currently taking. In what follows, our analysis of ESS data focuses on variations as to 'how much' mothers in particu- Joumal of European SocialFolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 26 Lewis et al. Table 3 Mothers' working hours and long-hours working, mothers and fathers1 Partnered 'mothers' Long-hours working With children aged 0- -15 With children aged 0- -6 Mothers Fathers Hours 'normally' Zero 1-19 20-34 35+ Zero 1-19 20-34 35+ 46+ 46+ worked including (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) overtime Austria 25.7 10.9 31.2 32.2 32.1 6.5 27.5 34.0 9.0 34.0 Belgium 22.4 9.3 31.7 36.6 22.6 8.8 31.4 37.3 9.9 33.2 Denmark 8.0 3.2 24.1 64.7 6.9 3.0 25.7 64.4 11.1 25.3 Finland 13.7 1.4 12.8 72.0 21.1 1.8 14.9 62.3 9.3 29.9 France 17.1 5.9 22.7 54.3 14.4 2.5 24.7 58.3 9.7 28.9 Germany 33.0 18.4 25.0 23.5 38.0 24.9 17.0 20.1 4.7 35.1 Great Britain 30.8 20.9 25.8 22.5 31.0 25.4 20.2 23.4 10.3 43.5 Greece 46.7 2.5 9.4 41.4 50.4 3.3 8.9 37.4 33.2 33.1 Ireland 39.4 11.8 22.3 26.5 34.0 12.2 22.6 31.2 4.9 42.1 Netherlands 21.1 33.3 34.1 11.5 21.9 28.8 38.5 10.8 1.9 27.7 Portugal 33.8 6.5 3.5 56.3 29.6 9.5 5.0 55.9 18.9 30.4 Spain 38.0 4.2 12.3 45.5 37.9 3.9 11.7 46.6 12.6 43.4 Sweden 9.4 2.4 25.8 62.4 12.8 0.9 26.6 59.6 8.3 24.1 Total 27.8 N = 3,002 12.8 22.6 36.8 29.2 N = 1,624 14.4 20.0 36.5 9.3 35.7 Noře: a The hours breakdown allocates those in paid work in 2003-05, many of whom are likely to be on formal or informal maternity or parental leave, according to the hours that they most recently worked. Long-hours figures are % of working mothers and fathers who work 46 hours or more, children aged 0-15. Sources: ESS and authors' calculations lar say they are employed, as to 'how much' child-care is also reported to be commodified, and as to people's preferences on hours of employment and formal child-care provision. It should be noted that the broad patterns of women's employment discussed in this section suggest that the development of labour markets in member states is also a very important determinant of demand for different kinds of reconciliation measures: if part-time jobs are relatively rare, as in Portugal or Greece, mothers may be less likely to perceive their options as including shorter hours of employment. round of data collection, which was carried out between late 2004 and early 2005. The sample analysed consists of 5,562 households comprising heterosexual couples and dependent children (0-15 years of age) where the female partner is aged 20-64 and one of the partners is the main respondent. In this article the terms 'mothers' and 'fathers' may include biological parents, stepparents, foster-parents or adoptive parents. The figures presented in the tables have all been weighted in order to account for the probability of selection into the sample.3 Findings from the ESS The data used are mostly drawn from the 'family, work and wellbeing' module of the European Social Survey (ESS), a cross-sectional study aimed to measure differences in social attitudes in Europe.2 We analyse the information that was collected in the EU-15 member states with the exception of Luxembourg and Italy (where the data on preferences are not robust). The data that we examine form part of the second Working hours Working hours (alongside state policies on services and leaves designed explicitly to 'reconcile' work and family responsibilities) are very important for balancing work and family. Working part-time has historically been the way that women in Western Europe have combined informal care for their children with employment. How much parents - more often women - work and to a lesser extent when Journal of European Social Policy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Patterns of paid and unpaid work in Western Europe 27 they work are both important determinants of the patterns of commodification of adults in households and of child-care. Our estimates of the average 'normal' full-time working hours reported by ESS mothers range from about 40 hours in Denmark, France and the Netherlands to 45 in Britain and 47 in Greece. This compares with a range of 44-9 hours for partnered fathers in the sample. Table 3 breaks down the ESS mothers' working hours and also shows the proportions of working mothers and fathers who work long (46+) hours per week. The countries in which the proportion of mothers working part-time is greater than the proportion working full-time (35+ hours) are Austria, Belgium, Germany, Britain, Ireland and the Netherlands. The long hours worked by a third of Greek mothers and 13-19 percent of Spanish and Portuguese mothers provide additional evidence for the Southern European long-hours culture. Taken with the relative lack of opportunity for part-time work, this may help to explain the lower female employment rates in these countries. At least a third of fathers report working 46+ hours per week in Austria, Belgium, Germany, Greece, Britain, Ireland and Spain; more than 40 percent in the last three countries. Even in the Nordic countries, around a quarter of fathers claim to have long working hours.4 Thus, where large proportions of mothers work part-time and/or where there is a low employment rate for mothers, more fathers tend to work long hours. However, in the Netherlands, where women tend to work very short part-time hours, the proportion of men working long hours is also relatively low. Indeed, Gornick and Heron (2006) have shown that, on average, US dual-earner married or cohabiting parents work 16 hours more per week than do Dutch parents. The Netherlands, then, has a short-hours culture for men as well as women. The proportions of fathers working long hours do not differ much in households where there are young children (except in Denmark). The breakdown of part-time hours for mothers in Table 3 shows important differences between countries. Many more Dutch mothers work under 19 hours than do those in Germany and Britain, though many fewer leave the labour market altogether. Part-time work in the Netherlands has long been decent work; traditionally less precarious than in Germany and especially Britain (Vasblom and Schippers, 2006). In most countries where there is a significant proportion of mothers working part-time they are working over 20 hours: indeed, this is now more true of Britain than was once the case (and is reflected in the FTE figures in Table 1). Table 3 also illustrates differences between Nordic countries for mothers with children under seven. In Denmark, fewer mothers stop paid work for long periods when children are small than in Finland or Sweden. In Finland, home-care leave extends well beyond our cut-off point for mothers' inclusion as paid workers (see note to Table 3) and in Sweden paid parental leave can be taken flexibly until the child is eight, enabling mothers there to provide more parental care at home (Moss and O'Brien, 2006). The ESS data permit the construction of households according to the working hours of the adults. Table 4 provides this information. Across the EU-13, nearly a third of households with children aged 0-15 are dual full-time households and the rest divide almost equally between male full-time/female part-time households (either long or short-hour part-time), and male single-earner households. Eurostat's 2003 data for the EU-25 (excluding Denmark, Ireland and Sweden) recorded about 35 percent of couples aged 20-49 with children under 12 as being dual full-time earners (Eurostat, 2005). Does this signal a shift from the one-and-a-half earner model to a dual full-time earner model? Caution is needed here. First, the issue is very susceptible to the definitions of full-time and part-time work. We have used 35+ hours to define full-time work, which corresponds to the maximum in French legislation (in 1998 and 2002) limiting working hours (Fagnani and Letablier, 2004); Eurostat (2005) used 30 hours. But ten years ago 40+ hours might have been deemed a more appropriate definition of 'full-time', which would have eliminated a lot of women whom we now define as full-time. Men's full-time working hours are still longer than those of women, and a father working 46 hours and a mother working 35 hours are both classified as full-time in Table 4. Second, earlier analysis did not always include countries where part-time work has not been strongly developed - Portugal, Spain, Greece (but Table 1 shows this is changing in Spain) - and where the one-and-a-half earner model never existed. Third, there is large variation among countries, including among 'other' forms of household. Fourth, although women's employment has risen in the last ten years, much of this has been part-time. Nevertheless, the fact that about a third of households with children now Journal of European SocialFolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 28 Lewis et al. Table 4 Household structure according to working hours, child 0-15 MFT + FFT1 MFT + F long PT MFT + F short PT ___________(%) 7.2 8.7 1.4 1.6 7.1 15.6 14.2 2.8 7.5 27.7 2.4 2.4 1.8 10.3 (%) (%) Austria 21.9 23.4 Belgium 29.3 22.9 Denmark 55.0 22.4 Finland 59.6 8.7 France 44.8 18.3 Germany 19.1 17.4 Great Britain 19.6 22.0 Greece 32.4 6.1 Ireland 22.2 17.5 Netherlands 6.3 27.5 Portugal 50.2 3.2 Spain 37.2 8.9 Sweden 53.8 24.8 Total 30.3 17.7 N = 5,412 Male sole earner* Other* (%) (%) 27.2 20.4 24.7 14.4 12.6 8.7 20.9 9.3 18.1 11.8 33.7 14.2 25.5 18.7 40.9 17.8 35.5 17.3 23.3 15.2 27.3 16.9 38.3 13.3 12.2 7.5 27.3 14.5 Notes: a MFT+FFT = both male and female partners normally working for 35+ hours a week; MFT + F long PT = male partner normally works 35+ hours, female 20-34 hours; MFT + F short PT = male partner normally works 35+ hours, female 1-19 hours; 'other' includes female sole earner or main earner (e.g. Portugal and Austria 9%), dual part-time household (e.g. Netherlands 7%), not in paid work (e.g. Greece 5%). This variable was constructed using information on working hours of the main respondent and what she/he said about her/his partner's working hours and counts the hours of mothers according to the hours of their most recent paid work in 2003-05. Sources: ESS and authors' calculations. have dual full-time earners is important. Particular countries weight this result heavily, especially the Nordic countries, but also France and Southern Europe. However, it should also be noted that EU enlargement has taken in many more dual full-time or male breadwinner households (Eurostat, 2005), which means that the one-and-a-half earner model is currently exceptional in the EU-27. To summarize broadly, we have: • Countries where dual full-time working is more usual than any other arrangement in couple households and represents a clear majority of partnerships with children: Denmark, Finland and Sweden. These countries have few single-earner couple households. The other Baltic countries also fall into this category though with less part-time work (Eurostat, 2005). • Countries that polarize between dual full-time earning and single-earner families, and where part-time work accounts for only a small proportion of couple households: Spain, Greece and Portugal, although the last of these has much higher dual full-time earner households than the other two. This pattern also tends to apply across Eastern Europe (Eurostat, 2005; see also Haas et al., 2006). • The rest are countries where the one-and-a-half model is or is soon likely to be more usual than any other: this is distinguished by substantial quantities of female part-time work, but nonetheless consists of a spectrum ranging from close-to-majority dual, full-time (France), to the Netherlands where well over half of households are one-and-a-half earners, and Ireland where male breadwinner families were still the largest single category in 2004, but with a substantial proportion of female part-time/male full-time households. So, just as we saw from Table 1, there is little evidence either of gender equality in terms of mothers' participation in paid work, or of convergence towards a dual full-time model outside the Nordic countries (and Portugal). It may be, as in Germany, that the total number of hours of paid and unpaid work for men and women are equal (Finch, 2006), Journal of European Social Policy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Patterns of paid and unpaid work in Western Europe 29 Table 5 Atypical working hours: 'mothers' and 'fathers' with children aged 0-15a Partnered 'mothers' Partnered 'fathers' Frequent Frequent Frequent overtime at Frequent Frequent overtime at Frequent evenings/nights short notice weekends evenings/nights short notice weekends (%) (%) (%) (%) (%) (%) Austria 16.4 19.4 29.5 12.4 15.3 26.2 Belgium 21.0 12.6 28.6 22.6 23.3 34.0 Denmark 17.8 13.9 29.0 23.4 18.4 30.5 Finland 32.2 11.6 28.1 32.1 26.6 33.9 France 16.6 15.6 29.3 26.9 36.1 29.9 Germany 12.2 17.1 24.2 22.3 36.9 28.3 Great Britain 24.4 10.7 21.3 36.9 30.4 50.0 Greece 13.6 6.0 40.2 17.0 16.1 43.9 Ireland 20.2 3.9 26.5 32.1 27.1 46.2 Netherlands 26.9 9.7 24.0 26.0 17.6 32.3 Portugal 10.9 9.0 22.4 19.4 19.3 40.1 Spain 16.7 7.3 26.6 26.8 24.0 39.8 Sweden 28.6 13.7 26.8 30.8 18.3 27.2 Total 18.4 N = 1,739 13.0 26.2 27.2 N = 2,254 29.2 36.2 Notes: a 'Frequent evenings/nights' and 'frequent overtime at short notice' = several times a week or every day; 'frequent weekends' = several times a month or every weekend. Question asked only of those who said their main activity was 'paid work'. Sources: ESS and authors' calculations. but this merely masks the huge gendered divisions which exist. Messenger (2004) has noted that data on hours show little change in average hours of work - indeed, there has been a slight fall in men's working hours - and yet people report a 'time squeeze', in large part because of women's increased labour market participation which means that each family unit is working more hours (see also Maclnnes, 2006). However, as Ellingsaeter (2006) has argued, it may be more appropriate to refer to this in terms of a 'care squeeze' rather than a 'time squeeze'. In many countries greater flexibility in terms of when they work is also being expected of employees, which may or may not benefit family life, depending on the amount of control that is permitted the worker. Table 5 shows when mothers and fathers work atypical hours, in the form of evenings and nights, overtime at short notice and weekends. As with long hours, more fathers than mothers do atypical working. The country rank orders are in many cases similar for men and women for the different types of atypical working, which may suggest that they are related to the nature of the labour market. A high percentage engaging in one form of atypical working - and here we may include long-hours working (see Table 3) - is often offset by low figures in other categories. Thus, Greece records high proportions working long hours and weekends, but relatively little by way of frequent evening working or short-notice overtime for mothers and fathers. Conversely, France records some of the highest figures for overtime at short notice for fathers. What distinguishes Denmark is that it records close-to-average or low figures for all these forms of working (Finland and Sweden have large percentages of mothers and fathers working evenings or nights). What distinguishes Britain is the extent to which high numbers of fathers not only work long hours, but also work all sorts of atypical hours, a finding in line with OECD data for all UK men (OECD, 2004: Tables 1.8 and 1A2.4). The OECD data for women's atypical working suggest it is more widespread in Britain than do the ESS data for mothers in Table 5. One possible explanation is methodological: that the ESS put the questions on atypical working only to those who said their main activity Journal of European SocialFolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 30 Lewis et al. was paid work, to the exclusion of shorter-hours part-time workers. Another is that reducing atypical work may be one way that British mothers seek to reconcile their work and care. However, the impact of atypical working may not be wholly negative since it permits a high degree of shift parenting (Lavalle et al., 2002), especially when, as in the UK, it is combined with a prevalence of short-hours working on the part of women. Using panel data, Warren (2003) showed that child-care in the UK is shared by 40 percent of working-class couples and 28 percent of middle-class couples (dual earner, manual worker couples are much more likely to work split schedules). It can also be argued that large numbers of men working long hours, together with a high incidence of other forms of atypical work for men, make it difficult for fathers to do care work. Nevertheless, research has shown that when these conditions do not prevail, fathers do not necessarily increase their share of unpaid work. Thus, Crompton (2006) attributes the high work/life conflict scores she found for women in France to the very unequal domestic division of labour. Working hours preferences Nearly ten years ago, in 1998, the European Foundation for Living and Working Conditions (Eurofound) conducted a large survey in the EU-15 and Norway to quantify the number of hours that women and men preferred to work, taking into account the need to earn a living (Atkinson, 2000; Bielenski et al, 2002; OECD, 2001). It concluded that the EU's strategy of bringing employment rates in Europe up to the US level was consistent with people's preferences but that, since most employees also wanted shorter working hours, 'the preference in Europe is for a combination of high labour market participation rates and short individual working times rather than the American combination of high employment rates and long working times' (Bielenski et al, 2002: Foreword). Already by 1998 employees were questioning the distinction between full-time and part-time work, with 'growing interest in a reformed or variable full-time norm located in the range of what actually constitutes "short" full-time and "long" part-time employment - i.e. around 30 hours' for women and 35 hours for men (Bielenski et al, 2002: Foreword, Chapter 4). Journal of European Social Policy 2008 18 (1) The data that we have analysed suggest that for working parents many of these conclusions still held six years later when the ESS fieldwork was carried out. Asked how many hours a week they would like to work 'bearing in mind that your earnings would go up or down according to how many hours you work',5 over 60 percent of working ESS fathers and half of working ESS mothers in our 13 countries said that they wanted to reduce their hours - fathers' preferred working hours were the lowest in Denmark, the Netherlands and Sweden (37-8 hours) and highest in Portugal (43 hours). For mothers the picture is more complicated. Country averages for preferred working hours ranged from about 20 hours for part-timers in Germany, Britain and the Netherlands to over 42 hours among Portuguese full-timers. Table 6 breaks down mothers' working hours preferences in more detail. Whereas over two-thirds of full-time mothers wanted to cut their working hours, over a third of part-time mothers wanted to work longer hours. Among full-timers, countries where the highest proportion of mothers were content with their current hours were Portugal (46%), Greece (40%) and the Nordic countries (29-31%). Countries where the highest proportion of full-time mothers wanted to cut their hours by 11 or more were the Netherlands, Germany, Britain and Belgium. Fewer mothers in Southern Europe wanted to reduce their hours, perhaps because household incomes are lower. However, preferences ranged widely, especially in relation to the hours actually worked at present (Table 3). A high proportion of full-time mothers in Spain wanted to reduce their hours, which is in line with increasing part-time work there (Table 1). A more detailed examination of ESS data on the three countries with a lot of part-timers (Germany, the Netherlands and the UK), presented in Table 3, suggests that short-hours part-time mothers in Germany (under 20 hours) were least content, with over half wanting to work more hours and a quarter wanting to work even less. The data suggest that over 60 percent of mothers in the Netherlands working short part-time hours want to work more. What may have changed, therefore, since the European Foundation's survey (although the ESS data are difficult to interpret) is the number of hours that most parents want to work and the extent to which there is interest in a long part-time 'norm' at all among mothers in the sense that analysts identified it from the 1998 responses. However, in line with Eurofound's survey, the ESS data suggest that Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Table 6 Difference between hours 'normally' worked and preferred working hours, mothers with children aged 0-15a Mothers in full-time work Mothers in part-time work Want less Want less Want more 11 hrs + Up to 10 hrs No change Want more 6 hrs + Up to 5 hrs No change Up to 10 hrs 11 hrs + (%) (%) (%) (%) Obs (%) (%) (%) (%) (%) Obs Austria 33.6 33.6 21.2 11.5 84 7.1 12.0 41.0 27.6 12.4 105 Belgium 38.8 37.3 19.4 4.5 67 13.9 8.3 44.4 16.7 16.7 72 Denmark 20.9 43.5 31.3 4.4 115 6.0 26.0 46.0 8.0 14.0 50 Finland 18.5 45.0 28.5 7.9 151 [0.0] [16.7] [20.0] [46.7] [16.7] 30 France 20.9 45.1 20.0 14.0 129 8.1 14.5 33.4 24.7 19.2 64 Germany 37.7 44.8 15.1 2.4 76 9.2 33.2 24.6 17.7 15.3 104 Great Britain [41.5] [30.8] [19.5] [8.2] 38 14.7 23.9 35.0 18.3 8.1 92 Greece 26.1 25.0 39.9 9.1 114 [2.5] [5.1] [5.1] [38.0] [49.4] 34 Ireland 33.9 39.7 21.0 5.3 80 7.4 13.8 37.6 25.3 16.0 96 Netherlands [40.7] [49.2] [10.2] [0.0] 26 9.3 17.8 35.3 29.8 7.9 162 Portugal 3.9 23.9 46.1 26.2 135 [0.0] [0.0] [0.0] [36.5] [63.5] 21 Spain 29.7 36.3 30.0 4.0 74 [3.5] [10.6] [41.6] [14.0] [30.3] 29 Sweden 14.0 52.7 31.0 2.3 129 8.6 10.3 44.8 22.4 13.8 58 Total 27.2 40.2 23.6 9.0 1,218 9.8 21.5 32.5 21.3 15.0 917 Note: a Partnered mothers in paid work at any time from 2003 to survey date, most recent actual hours. [ ] = Sample size under 50 Sources: ESS and authors' calculations. Table 7 Child-care use and formal child-care preferences among mothers in paid work, child aged 0-6a Types of child-care used for youngest child Mother's views of formal child-care provision Parent Other Want Want Provision is Formal at home Grand-parents unpaid Other 'much more' 'slightly more' 'about right' Less (%) (%) (%) (%) (%) Obs (%) (%) (%) (%) Obs Austria 26.6 13.7 41.9 16.1 1.6 51 5.8 17.8 76.4 0.0 72 Belgium 37.3 7.5 47.8 6.0 1.5 67 5.5 15.1 75.3 4.1 73 Denmark 67.8 8.1 16.1 5.8 2.3 87 2.3 9.0 84.3 4.5 89 Finland 60.9 10.9 23.4 1.6 3.1 64 0.0 11.2 74.2 14.6 89 France 67.8 15.6 10.7 0.0 6.0 94 15.3 37.1 47.7 0.0 102 Germany 45.1 19.8 32.2 0.7 2.3 74 7.3 17.8 74.4 0.6 89 Great Britain 30.3 26.4 30.9 10.9 1.6 61 2.6 15.5 79.9 2.0 73 Greece 26.1 8.7 61.7 3.5 0.0 52 11.2 25.1 62.0 1.7 82 Ireland 40.2 11.0 31.7 17.1 0.0 92 5.4 21.0 71.0 2.6 106 Netherlands 32.5 22.3 34.5 9.6 1.0 98 1.0 10.4 86.7 1.9 105 Portugal 43.8 9.5 33.6 10.4 2.7 90 36.1 37.6 26.2 0.0 103 Spain 32.5 19.4 30.1 11.0 7.0 46 16.0 29.1 53.2 1.7 57 Sweden 69.1 21.0 6.2 2.5 1.2 81 2.2 7.7 83.5 6.6 91 Total 46.8 18.4 26.2 5.2 3.4 957 9.6 23.1 65.7 1.5 1,131 Notes: a 'Parent at home' = no child-care needed (e.g. don't go out to work, always one parent at home, I/my current partner care for children); 'Other unpaid' = ex-spouse, ex-partner, other family member and other unpaid care at carer's or child's home (e.g. neighbours). The data on usage are for mothers in paid work within the last seven days. The data on views of child-care provision also include mothers in paid work any time in 2003-05. Sources: ESS and authors' calculations. Patterns of paid and unpaid work in Western Europe 33 working-time preferences vary considerably between EU member states. In addition, there is still a gap between mothers' and fathers' preferred working hours, which raises doubts as to how welcome a gender-equal worker/carer model family would prove to be (Gornick and Meyers, 2003). As discussed above, the preferences articulated in this and other surveys may be constrained or affected by factors such as education, the options that the labour market offers, level of child-care provision or a culturally determined sense of the 'ought'. But there are perhaps three main overriding messages from these data: fathers want to work much less; mothers want to be employed, for the most part long part-time or full-time hours; and in most countries mothers want to have a wide choice of hours, presumably to fit their different and changing domestic situations. Above all, it is clear that working hours remain an important part of the 'reconciliation agenda' both in respect of practices and preferences. The use of formal and informal child-care It might be expected that in countries with high proportions of mothers employed full-time, the use of formal child-care would be high. Table 7 sets out the responses of working mothers with a youngest child aged 0-6 to a question asking about the usual type of child-care that was used in their household. This table shows that formal child-care is hugely important in the Nordic countries and France, where female employment is high, and there are many dual full-time households together with a long tradition of formal child-care provision that is accessible, affordable and relatively high quality. Care by the mother or partner is particularly important in Britain, where fathers working atypical hours provide significant amounts of care (see above). However, figures such as those in Table 7 may hide substantial complexities. For example, further analysis shows that usage of formal care in Finland and Sweden is about twice as high for pre-school children over three years of age as it is for babies and toddlers, while in Denmark formal care is used more from a younger age. Nearly half of parents of 0-2-year-olds in Sweden and a third in Finland, but under 20 percent in Denmark, said that their youngest child was cared for by a parent (or partner) at home.6 Grandparents (often grandmothers) are a highly significant source of child-care everywhere, except in Sweden, Denmark and France. In the first two, a high proportion of women aged 55-64 (67% and 53% respectively) are in employment, which may help to explain the lack of grandparental care, but this is not the case in France (37%) (OECD, 2005). Wheelock et al. (2003) draw attention to the importance of redistribution between the generations in respect of care and their concept of the 'family earning household' widens household boundaries to include grandparents' 'complementary economic activity'. Finch (2006) reported that mothers in employment in the UK and in Sweden do not provide more child-care for young children than mothers of all-age children. She suggests this is because in Sweden they call upon formal care, while in the UK they increase their use of informal care. The data on child-care usage in Table 7 are restricted to young children of mothers who were in paid work within the previous seven days because these mothers are likely to need child-care most. The responses from other ESS mothers and from ESS fathers support the general picture presented in Table 7. Table 7 also shows the preferences of working mothers whose youngest child is under seven for formal child-care provision: and here we include responses from mothers who were in paid work up to two years before the survey - to include the child-care views of mothers on statutory or nonstatutory maternity and parental leave. In most countries, a substantial majority are content with the amount of formal child-care that is available. The exceptions are Portugal, where a high percentage of mothers wanted 'much more' child-care, and, to a lesser extent, France and Spain. Portugal has a relatively high rate of mothers employed full time and of dual full-time earner households, and, given this, relatively low levels of formal child-care provision (Plantenga and Remery, 2005). In the case of Spain, the employment of child-care workers (often migrants) by households has become increasingly important (Flaquer and Navarro, 2005) and part-time work is growing, but the fact that the women who are in the labour market tend to work full-time (and are concentrated in the younger age groups) helps to account for the wish for more formal provision. France, however, has relatively good provision of formal child-care. Dissatisfaction in this country may relate to the changes in child-care policy, moving away from subsidy of provision towards subsidy of the household wishing to employ a child-care worker (as a labour activation measure) (Fagnani, 1998) Journal of European SocialFolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 34 Lewis et al. and/or to relative lack of participation by men in domestic work. It would, of course, be useful to know to what extent enthusiasm for formal provision is impeded by current costs, quality, availability, short opening hours etc.; de Henau et al. (2006) have stressed the extent to which these factors affect take-up. Not shown in Table 7 are the views of non-working mothers and fathers. Most ESS mothers who were not working also said that child-care provision was 'about right' (except in Portugal).7 We note that the responses of ESS fathers who chose to reply to the questions on child-care (both usage and demand for formal care) were not very different from mothers in the same country (even though the male main respondents were from different households). Apart from Portugal, France and Spain, there is no evidence in these data that ESS mothers would at this moment in time support expensive policies which provided universal child-care as a 'public good' (Folbre and Bittman, 2004) in countries that do not already have it. However, as with the working hours preferences which respondents to the survey expressed, the views of mothers on child-care may owe much (or little) to constraints, structural factors or culturally determined attitudes that may change: not shown in Table 7 is the fact that demand for more child-care provision appears to be slightly stronger among ESS mothers with the lowest and the highest educational qualifications -perhaps because the former need to work more for financial reasons and the latter want to work more. The ESS survey did not examine mothers' attitudes to the relationship between formal child-care and the health and developmental needs of children -inevitably also a determinant of public policy. Policy implications We confine our comments to the implications of our findings for policies to meet the challenge of combining employment and care work, particularly the current EU-level emphasis on the importance of formal child-care provision (always acknowledging the large issue of how far which policies are likely to matter in which countries). 'Reconciliation' policies may take many forms; the ESS data provide information only on patterns and preferences in respect of working hours and on child-care provision. The survey did not ask about preferences for leaves - a very important means of decommodifying labour, the Journal of European Social Policy 2008 18 (1) take-up of which is very difficult to measure (Bettio and Plantenga, 2004; Ellingsaeter and Leira, 2006; Hobson et al, 2006). Indeed, it is possible to define care leaves as necessary modernization of the traditional (male) career model (Pfau-Effinger, 2004), and there is evidence in Germany and the Netherlands that mothers prefer parental care (Knijn and Ostner, forthcoming), which points to the importance of parental leave policies. With regard to patterns of commodification, the vast majority of EU-13 countries are very far from full individualization in terms of employment. In addition, the provision of unpaid child-care remains important. There is a care squeeze for households as the employment rates of women rise, especially as their hours of work also lengthen, and given that in many countries over a third of fathers with young children work long hours. EU-level policy assumes that more commodification of child-care will promote labour force participation. The Portuguese case shows that relatively high, full-time employment for mothers can occur without equivalently high provision of child-care, but the very high proportions of Portuguese women wanting more formal child-care provision indicates that this situation is hardly popular. Grandparents are shown to be a very important source of child-care. Yet EU policy on 'active ageing' is for grandparents to engage in the labour market (EC, 2002). If this were to happen, as is already the case in Sweden, then it is likely that the demand for child-care would soar - with huge budgetary implications. Given that in most countries mothers with a youngest child under seven express a preference for either the amount of formal child-care provision that already exists, or only slightly more, the role of grandparents as care providers seems to warrant more recognition in policy development.8 In any case, the figures in Table 7 suggest that work/family reconciliation measures need to extend beyond a simple pressure for more formal provision which characterizes current EU-level policy. There do seem to be relatively strong preferences for changes in working hours, with the ESS data supporting previous findings from Eurofound (Atkinson, 2000; Bielenski et al, 2002). This provides support for the development of policy packages that include permission for parents to decommodify their labour via child-care leaves,9 and entitlements to part-time or flexible patterns of work, as well as formal child-care. At the EU level, considerable attention was paid to the development of parental leave policies in the mid- Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Patterns of paid and unpaid work in Western Europe 35 1990s, and more recently attention has shifted to promoting the development of formal child-care. The motivation has been instrumental: mainly to promote female labour market participation. The issue of working hours has been raised under the heading of work/family reconciliation policies (EC, 2006), but has been given much less attention and mostly restricted to the issue of the 48-hour working time limit (a figure far above most parents' maximum hours). As a comparison between the FTE figures for Greece and the Netherlands shows (Table 1), the significance of female employment for achieving the Lisbon Strategy's economic goals depends not only on increasing the numbers of women working but also the hours they work. For this, employers must be persuaded or even required to make it easier for mothers to choose a much wider range of working hours options. In this respect, the Commission's launch of its new preliminary consultation in 2006 may signal a change. It is long overdue: the patterning of working hours in terms of length and flexibility over the life-course is a complicated and urgent issue (Anxo and Boulin, 2006; Eurofound, 2006; Schmid, 1998). A narrow policy focus is unlikely to be the answer. Countries vary enormously in terms of the nature of the existing policy package and patterns of adult labour market participation. Portuguese women express a strong preference for much more formal child-care; Dutch, German and British women are relatively satisfied with the amount they have, despite having much less developed formal provision than the Nordic countries. As matters stand, member states tend to have emphasized one element of the policy package more than others (de Henau et al., 2006), which has to do with different cultures of work and care, and 'ideals of care' (Kremer, 2006). Thus, different patterns of development in the work/family balance policy field are likely to be needed in different countries. Respect for parental choice is increasingly an issue with regard to the gender divisions of unpaid care work and employment, and the degree of commodification that is achieved, although the notion of choice may be distorted by political rhetoric (Ellingsaeter and Leira, 2006; Lewis, 2007). Norms and values matter in policy making, especially in family policy (Strohmeier, 2002); it remains a problem as to how far the state should lead or follow. It is in any case particularly difficult for the European Commission to advocate a particular policy approach for all member states. Acknowledgements We gratefully acknowledge support of ESRC grant number 225 25 2001 and the helpful advice of Rory Fitzgerald of ESS. Notes 1 Interestingly, Cyprus has already met the Lisbon target. 2 In the UK, responses cover only GB, i.e. excluding Northern Ireland. 3 The weights used do not adjust for non-response, i.e. if non-respondents do not match the probability of selection into the sample, their absence will distort the sample (see ESS Sampling Report). The ESS achieved its 70% target response rate in very few countries and rates were under 60% in France (43.6%), the UK (50.6%), Germany (51.0%) and Spain (54.6%). Almost all the data we have used have non-response rates for individual questions of under 10% and usually much lower. However, for these and other reasons, the ESS responses are not always in line with the information derived from the much larger surveys like Eurostat. 4 We have relied on respondents' answers to the ESS, but it is interesting that the proportion of female respondents reporting that their male partners work 46+ hours is lower - sometimes much lower - than the proportion of male respondents shown in the table (except in Greece where it is much higher). 5 This is important given Jacobs and Gerson's (2004) findings for the US: that while most people - men and women - want to work and care, they do not want to cut their working hours if that means losing money. 6 Our data were not robust enough for other countries to distinguish formally between child-care used by parents of children above and below the age of three. 7 Strictly, the ESS questionnaire enquired about views of child-care provision 'in your present situation'. We find it hard to believe that so many mothers, working or not, would have expressed themselves content with current levels of provision if a wish to change their 'present situation' had made them strongly dissatisfied. 8 From the point of view of grandparents themselves, there is the issue of how far they want to care for grandchildren. 9 Our work addresses child-care only. Elder-care is just as important and will become more so with population ageing. References Akerloff, G. A. and Kranton, R. E. (2000) 'Economics and Identity', The Quarterly Journal of Economics 115 (3): 715-53. Anxo, D. and Boulin, J-Y. (2006) 'The Organisation of Time over the Life-course: European Lrends', European Societies» (2): 319^1. Atkinson, J. (2000) Employment Options and Labour Market Participation. Dublin: European Foundation for the Improvement of Living and Working Conditions. Journal of European SocialPolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. 36 Lewis et al. Becker, P. E. and Moen, P. (1999) 'Scaling Back: Dual-earner Couples' Work-Family Strategies', Journal of Marriage and the Family 61 (4): 995-1007. Bettio, F. and Plantenga, J. (2004) 'Comparing Care Regimes in Europe', Feminist Economics 10 (1): 85-113. Bielenski, PL, Bosch, G. and Wagner, A. (2002) Working Time Preferences in Sixteen European Countries. Dublin: Eurofound. Bonoli, G. (2005) 'The Politics of the New Social Policies: Providing Coverage against New Social Risks in Mature Welfare States', Policy and Politics 33 (3): 431-49. Crompton, R. (2006) Employment and the Family. Cambridge: Cambridge University Press. Crompton, R. and Lyonette, C. (2006) 'Some Issues in Cross-national Comparative Research Methods: a Comparison of Attitudes to Promotion, and Women's Employment, in Britain and Portugal', Work, Employment and Society 20 (2): 389-400. Daly, M. (2002) 'A Fine Balance: Women's Labor Market Participation in International Comparison', in F. W. Scharpf and V. A. Schmidt (eds) Welfare and Work in the Open Economy, Vol. II, pp. 467-510. Oxford: Oxford University Press. de Henau, J. Meulders, D. and O'Dorchai, S. (2006) 'The Childcare Triad? Indicators Assessing Three Fields of Child Policies for Working Mothers in the EU-15', Journal of Comparative Policy Analysis 8 (2): 129-48. EC (2002) Increasing Labour Force Participation and Promoting Active Ageing. COM (2002) 9 final. Brussels: European Commission. EC (2006) First Stage Consultation of European Social Partners on Reconciliation of Professional, Private and Family Life, SEC(2006)1245. Brussels: European Commission. Ellingsaeter, A-L. (2006) 'The Norwegian Childcare Regime and its Paradoxes', in A-L. Ellingsaeter and A. Leira (eds) Politicising Parenthood in Scandinavia. Gender Relations in Welfare States. Bristol: Policy Press. Ellingsaeter, A-L. and Leira, A. (eds) (2006) Politicising Parenthood in Scandinavia. Gender Relations in Welfare States. Bristol: Policy Press. Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Eurofound (2006) Working Time Options over the Life Course: New Work Patterns and Company Strategies. Dublin: Eurofound. European Social Survey (2005) Round 2. Available at: [http://ess.nsd.uib.no]. Eurostat (2005) Gender Gaps in the Reconciliation between Work and Family Life. Statistics in Focus 4/2005. Brussels: Eurostat. Fagan, C, Grimshaw, D. and Rubery, J. (2006) 'The Subordination of the Gender Equality Objective: the National Reform Programmes and "Making Work Pay" Policies', Industrial Relations Journal 37 (6): 571-92. Fagnani, J. (1998) 'Helping Mothers to Combine Paid and Unpaid Work - or Fighting Unemployment? The Ambiguities of French Family Policy', Community, Work andFamily 1 (3): 297-312. Fagnani, J. and Letablier, M-T. (2004) 'Work and Family Life Balance: the Impact of the 35-hour Laws in France'. Work, Employment and Society 18 (3): 551-72. Journal of European Social Policy 2008 18 (1) Finch, N. (2006) 'Gender Equity and Time Use: How do Mothers and Fathers Spend their Time?', in J. Bradshaw and A. Hatland (eds) Social Policy, Employment and Family Change in Comparative Perspective. Cheltenham: Edward Elgar. Finch, J. and Mason, J. (1993) Negotiating Family Responsibilities. London: Tavistock/Routledge. Flaquer, L. and Navarro, L. (2005) 'Change in Family Structures and Informal Care Work', paper given at the 'Hidden Work Regimes' conference (May), Hamburg. Folbre, N. and Bittman, M. (eds) (2004) Family Time. London and New York: Routledge. Gershuny, J. and Sullivan, O. (2003) 'Time Use, Gender and Public Policy Regimes', Social Politics 10 (2): 205-28. Gornick, J. and Heron, A. (2006) 'The Regulation of Working Time as Work-Family Reconciliation Policy: Comparing Europe, Japan, and the United States', Journal of Comparative Policy Analysis 8 (2): 149-66. Gornick, J. and Meyers, M. (2003) Families that Work. Policies for Reconciling Parenthood and Employment. New York: Russell Sage. Haas, B., Steiber, N, Hartel, M. and Wallace, C. (2006) 'Household Employment Patterns in an Enlarged European Union', Work, Employment and Society 20 (4): 751-71. Hakim, C. (2000) Work-Lifestyle Choices in the Twenty-first Century: Preference Theory. Oxford: Oxford University Press. Himmelweit, S. (2005) 'The Determinants of Caring Behaviour', Full Report of Research Activities and Results. Swindon: ESRC. Himmelweit, S. and Sigala, M. (2004) 'Choice and the Relationship between Identities and Behaviour for Mothers with Pre-school Children: Some Implications for Policy from a UK Study', Journal of Social Policy 33 (3): 455-78. Hobson, B. (2004) 'The Individualised Worker, the Gender Participation and the Gender Equity Models in Sweden', Social Policy and Society 3(1): 75-83. Hobson, B., Duvander, A-Z. and Hallden, K. (2006) 'Men and Women's Agency and Capabilities to Create a Worklife Balance in Diverse and Changing Institutional Contexts', inj. Lewis (ed.) Children, Changing Families and Welfare States. Cheltenham: Edward Elgar. Jacobs, J. A. and Gerson, K. (2004) The Time Divide. Work, Family and Gender Inequality. Cambridge, MA: Harvard University Press. Knijn, T. and Ostner, I. (forthcoming) 'The "Meaning" of Children in Dutch and German Family Policy', in A. Leira and C. Saraceno (eds) Comparative Social Research 2008. Kok, W. (2004) 'Report from the High Level Group'. Luxembourg: OOPEC. Kremer, M. (2006) 'How Welfare States Care: Culture, Gender and Citizenship in Europe', unpublished PhD thesis, University of Utrecht. Lavalle, I., Arthur, S., Millward, C, Scott, J. and Clayden, M. (2002) Happy Families? Atypical Work and its Influence on Family Life. Bristol: Policy Press. Leitner, A. and Wroblewski, A. (2006) 'Welfare States and Work-Life Balance', European Societies 8 (2): 295-317. Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution. Patterns of paid and unpaid work in Western Europe 37 Lewis, J. (2001) 'The Decline of the Male Breadwinner Model: the Implications for Work and Care', Social Politics 8 (2): 152-70. Lewis, J. (2002) 'Gender and Welfare State Change', European Societies 4 (4): 331-57. Lewis, J. (2006) 'Employment and Care: the Policy Problem, Gender Equality and the Issue of Choice', Journal of Comparative Policy Analysis 8 (2): 103-14. Lewis, J. (2007) 'Families, Individuals and the State', in J. Hills, J. Le Grand and D. Piachaud (eds) Making Policy Work. Bristol: Policy Press. Maclnnes, J. (2006) 'Work-Life Balance in Europe: a Response to the Baby Bust or Reward for the Baby Boomers?' European Societies 8 (2): 223-49. McRae, S. (2003) 'Constraints and Choices in Mothers' Employment Careers: a Consideration of Hakim's Preference Theory', British Journal of Sociology 54 (3): 317-38. Messenger, J. C. (ed.) (2004) Working Time and Workers' Preferences in Industrialized Countries. Finding the Balance. London: Routledge. Moss, P. and O'Brien, M. (eds) (2006) International Review of Leave Policies and Related Research, Employment Relations Research Series No. 57. London: Department of Trade and Industry. OECD (2001) Employment Outlook. Paris: OECD. OECD (2004) Employment Outlook. Paris: OECD. OECD (2005) Employment Outlook. Paris: OECD. Pfau-Effinger, B. (2004) 'Welfare State Policies and the Development of Care Arrangements', European Societies 7 (2): 321-47. Plantenga, J. and Remery, C. (2005) Reconciliation of Work and Private Life: a Comparative Review of Thirty European Countries. Brussels: European Communities. Schmid, G. (1998) Transitional Labour Markets: a New European Employment Strategy, WZB Discussion Paper FS 1 98-206. Berlin: WZB. Strohmeier, K. P. (2002) 'Family Policy - How does it Work?', in F-X. Kaufman, A. Kuijsten, H-J. Schulze and K. P. Strohmeier (eds) Family Life and Family Policies in Europe, Vol 2. Oxford: Oxford University Press. Vasblom, J. D. and Schippers, J. (2006) 'Changing Dynamics in Female Employment around Childbirth: Evidence from Germany, the Netherlands and the UK', Work, Employment and Society 20 (2): 329-47. Visser, J. (2002) 'The First Part-time Economy in the World: a Model to be Followed?' Journal of European Social Policy 12 (1): 23-42. Wallace, C. (2002) 'Household Strategies: their Conceptual Relevance and Analytical Scope in Social Research', Sociology 36 (2): 275-92. Warren, T. (2003) 'Class- and Gender-based Working Time? Time Poverty and the Division of Domestic Labour', Sociology 37 (4): 733-52. Wheelock, J., Oughton, E. and Barris, S. (2003) 'Getting by with a Little Help from your Family: Towards a Policy-relevant Model of the Household', Feminist Economics 9 (1): 19-45. Journal of European SocialFolicy 2008 18 (1) Downloaded from http://esp.sagepub.com at Uppsala Universitetsbibliotek on April 25, 2008 © 2008 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.