Labour Supply in the UK

Labour Supply in the UK Second report for the project WELFARE POLICY AND EMPLOYMENT IN THE CONTEXT OF FAMILY CHANGE. Naomi Finch Social Policy Research Unit University of York Heslington York Y010 5DD Tel: 01904 432626 E-mail: nlf1@york.ac.uk

Introduction After recession at the very beginning of the 1990s, the UK experienced economic growth. But more recently the UK economy has begun to slow. Despite this, however, the labour market has remained surprisingly resilient, with continued increased aggregate employment. Political change from a Conservative to a Labour government in 1997 has brought with it significant changes regarding attitudes towards labour market behaviour. The New Labour government has brought with it an emphasis on child welfare, especially in relation to the eradication of child poverty. This has meant an emphasis on employment for all and making work pay, but especially for those responsible for children. The most notable outcomes have been the integration of mothers with young children into the labour market and the continued demise of the breadwinner family. Lone parents have also been encouraged to enter the labour market, although their still remains significant barriers to their integration. Whilst women have become more integrated into the labour market, they still remain the main carers, and the nature of their labour market participation reflects this. Men, on the other hand, are losing out on family life due to a long hours culture. New Labour has begun to recognise this gender difference, and part of this paper gives an overview of the current labour market gender equity policies, although a full analysis can be found in the third report on policy. Overall, this report aims to investigate men and women s experiences of the labour market, and the effect that caring has upon their work. It shall explore the labour market behaviour of men and women as individuals, as well as that of families and households. It will also explore inputs (education) and outcomes of labour market activity (earnings and poverty). 1 Labour market participation, employment and unemployment The figures below show the labour market participation rates, employment rates and unemployment rates for men and women between 1992 and 2002. Whilst men s participation rates have marginally declined, a gradual increase in female participation rates can be observed, resulting in a move towards a convergence of men and women s participation rates. If we look at male and female employment (figure 2) and unemployment (figure 3) rates, we can find an explanation for these trends. The female employment rate has increased by 4 percentage points between 1992 and 2002, but the equivalent unemployment rates are down by only 3 percentage points, resulting in the observed increase in female labour market participation levels. However, whilst the male employment rate has increased at a 2.5 percentage points, their unemployment rate has fallen at a steeper rate (5.9 per cent), resulting in the observed gradually decreasing male participation levels. Whilst there was a significant gap between male and female unemployment rates at the beginning of the 90s, there was little difference between the two in 2002. The unemployment rate for both sexes now shows some signs of increasing.

Figure 1: participation rates of men and women 90 85 80 75 70 65 % 60 Males aged 16-64 Females 16-59 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Years Source: LFS Spring quarters March- May Seasonally adjusted Figure 2: Employment rates of men and women 85 80 75 70 65 % 60 Males aged 16-64 Females 16-59 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 year Source: LFS Spring quarters March- May Seasonally adjusted unpaid family workers included

Figure 3: Male and female unemployment rate 14 12 Males aged 16-64 Females 16-59 10 8 6 4 2 % 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 year Source: LFS Spring quarters March- May Seasonally adjusted Unemployment varies according to age for both men and women. Figures 4 shows the unemployment rates for men and Figure 5, for women. Since 1994, all age groups for both sexes have experienced a general downward trend in unemployment, with the exception of the youngest age group (aged 16-17), whose position has not improved between 1994-2002. The vulnerability of this group is apparent for both sexes, although the actual unemployment rate is persistently higher for men than for women. The relatively high unemployment of men and women aged 16-17 is a result of a combination of limited labour market experience coupled with low-level qualifications, due to their young age. Both place this age group in a disadvantaged position in the labour market.

Figure 4: Male unemployment (by age) 25 20 15 10 5 % 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 year 16-17 18-24 25-34 35-49 50-64 65+ Source: LFS Spring quarters March- May Seasonally adjusted F igure 5: Female unemployment (by age) 25 20 15 10 5 % 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 year 16-17 18-24 25-34 35-49 50-59 60+ Source: LFS Spring quarters March- May Seasonally a djusted 2 Dependent children and labour marke t participation H akim (1996) has demonstrated that married women tend to enter the labour market after their youngest child has reached school age, leading to an M-shape d labour market profile for British women. There is a high level of labour market participation by the younger age groups followed by a fall as women reach child bearing age wh en they break from their jobs to look after children, followed by another increase from

their mid thirties onwards. In contrast, the typical male pattern is a continuous and full-time labour market attachment from leaving full-time education until retirement (Hakim, 1996). However, there is evidence that the pattern of mothers interrupting employment while children are below school age children is becoming less dominant. First, the M- shaped profile is becoming less of women s employment is less marked. Figure 6 shows labour market participation according to age of men and women. In 1992, the M-shaped profile for women is apparent but by 2002, female labour market participation demonstrates a gradual increase until women reach the age of 35-49; their profile is closer to that of men. Figure 6: Participation rates of men and women by age in 1992 and 2002 100 90 80 70 60 50 40 30 20 10 Women: 1992 Women: 2002 Men: 1992 Men: 2002 % 0 16-17 18-24 25-34 35-49 50-59/64 :60/65 Age and over Source: LFS Spring quarters March- May Seasonally adjusted Second, the increase in female economic activity rates during the 1990s has largely been due to the increased economic activity of women with dependent children, especially those with pre-school children. Between 1991 and 2001, the economic activity rate for women with pre-school children increased from 48 per cent to 57 per cent (table 1). However, for women without dependent children the economic activity rate in 2001, at 75 per cent, was the same as in 1991. Women on maternity leave are classified as in employment, so this rise indicates that an increasing number of women return to the labour market within a shorter period after the birth of their children than previously. This can be attributed to the statutory right to retain ones job after maternity leave, which has encouraged women to return to work; in 1996, 67 percent of mothers had returned to work within 9 months of the child s birth compared to 45 percent in 1988 (Callender et al, 1997). Also, it suggests

that an increasing number do not leave the labour market at all while having their children. This is emphasised by the decrease in the proportion of mothers with preschool children who were looking after family and home, especially for those with pre school children; from 47 per cent in 1991 to 38 per cent in 2001 (table 1). Table 1: Economic activity status of women 1 : by age of youngest dependent child, 1991 and 2001 Age of youngest dependent child Under 5 5-10 11-15 16-18 2 No dependent children Employed (1991) 42 66 73 75 70 65 Employed (2001) 54 70 75 80 72 69 Unemployed (1991) 6 6 4 3 5 5 Unemployed (2001) 3 3 4 2 3 3 All Economically active (1991) 48 71 77 78 75 70 All Economically active (2001) 57 73 79 83 75 72 Looking after family/ home (1991) 47 22 15 13 6 17 Looking after family/ home (2001) 38 18 12 7 4 13 Students (1991) 1 1.... 6 4 Students (2001) 1 2 1.. 8 5 Other inactive (1991) 4 5 7 8 11 9 Other inactive (2001) 3 6 8 10 13 10 All economically inactive (1991) 51 28 22 21 23 29 All economically inactive (2001) 43 27 21 17 25 28 1 Aged 16-59. At Spring each year. 2 Those in full-time education. 3 Based on the ILO definition. See Appendix, Part 4: ILO Unemployment. Source: Labour Force Survey, Office for National Statis tics So we have seen that mothers with dependent children, especially those with very young children, have in the last ten years become more integrated into the labour market. However, despite these changes the presence of a young child in the family still has a significant effect on the economic activity of women (more than men). First, mothers with young children are still less likely to be employed than mothers with older children, with employment and participation rates increasing as the age of the youngest child gets older (table 1). Also, there is still some evidence that women, but not men, are delaying labour market participation until their children are older, since as we have seen in figure 6 their participation does not peak until age 35-49, whilst men s labour market participation reaches a peak at age 25-34. Indeed, table 2 demonstrates that whilst women with young children are less likely to be in employment than all women, men with young children are more likely to be employed. For women with children under 2, only 50 percent are in employment; 20 percentage points lower than the rate for all women. However, 90 per cent of men with children under two are employed; ten percentage points more than all men. All

These figures are similar for women and men with children aged 2-4. Employment rates for women increase with the age of the youngest child, but those for men decrease. Many studies have attributed these differences in employment to the fact that women bear children and disproportionately spend more time in caring for those children than men (Crompton, 1997; Iacavou and Berthoud, 2000). Table 2: Proportion of people in employment by age of youngest child, UK Women Men ALL All people of 70 80 75 working age (% in employment) Under 2 years 50 90 69 2-4 years 55 89 71 5-9 years 68 87 76 10-15 years 72 80 76 16-18 years in full time education 67 69 68 Base: All people of working age (men 16-64, women 16-59) in employment Source: Estimates form the Labour Force Survey (Autumn, 2002) Lone parents and labour market participation The presence of dependent children has a dramatic effect on the economic activity rates of lone parents. As a result, New Labour have set a target to increase lone p arent employment to 70 per cent by 2010. In 2002 the lone parent activity rate was significantly lower than for the female working age population as a whole (57 per cent, compared with 73 per cent in 2002) 1. The gap, however, is decreasing; over the past decade, more lone mothers have become economically active, rising by 9 percentage points since 1992, as shown in table 3. The increase has been largest for those with pre-school aged children, reflecting the general upward trend for women to return to the labour market sooner after the birth of their children. As with all mothers, the economic activity rate of lone mothers increases with the age of the youngest child from 39 per cent for those with a child under 5, to 80 per cent for those whose youngest dependent c hild is aged 16 to 18. In spring 2002, 12 per cent of all lone parents were male. Their economic activity rate is higher than female lone parents; 77 per cent compared with 57 per cent, and is only 7 percentage points lower than the economic activity rate for all men of working age. This demonstrates that, as with all mothers and fathers, the presence of dependent children affects lone fathers very differently than it does lone mothers. 1 The underlying reasons for this are discussed below, in section 4 on Non-employment and workless households.

Table 3: Economic activity status of lone mothers: 1 by age of youngest dependent child, 1992 and 2002 United Kingdom Age of youngest dependent child Under 5 5-10 11-15 16-18 2 All Employed (1992) 21 44 61 63 38 Employed (2002) 34 55 64 74 51 Unemployed 3 (1992) 8 12 6.. 9 3 Unemployed (2002) 5 8 7.. 6 All Economically active (1992 29 55 67 72 48 Economically active (2002) 39 62 70 80 57 Looking after family/ home (1992) 65 35 19.. 43 Looking after family/home (2002) 52 25 15.. 31 Other inactive (1992) 6 10 14 17 9 Other inactive (2002) 9 13 15 16 12 All economically inactive (1992) 86 68 62 50 73 All economically inactive (2002) 61 38 30 16 43 1 Females aged 16-59. At spring each year. These estimates are not seasonally adjusted and have not been adjusted to take account of the recent Census 2001 results. 2 Those in full time education. 3 ILO definition Source: Labour Force Survey, Office for National Statistics 3 A-typical or flexible work forms The increase in women s participation in the labour market is a result of the increase in A-typical or flexible types of work, mainly part-time work, rather than an increase in full-time employment (Hakim, 1996; Crompton, 1997; Payne, 1991). A report conducted by the National Citizens Advice Bureau (NACAB 2001) points out that flexible work forms means something different to different people and therefore the expression does not have a single definition. It states that for some, flexibility is about adaptability and employability whilst for others flexible is synonymous with deregulation, so the main feature of flexibility is the refusal to accept responsibility for the workforce, and the absence of employment rights (NACAB 2001:1.3). Indeed, flexible and atypical working can be a solution to different problems, depending whose perspective is taken into consideration: Governments see it as a solution to rising unemployment, and encouraging people, especially women into the labour market; employers see it as a way of securing flexibility, and perhaps cheap labour; and employees see it as a way to combine paid and unpaid work (Humphries and Rubery 1995). Non-regular forms of employment in Britain can be divided into three categories: part time work, temporary work and other forms of flexible-working patterns work. We discuss these below. Part-time work Comparative studies have demonstrated that the shift to part-time work is most dramatic in the UK compared to other countries (Bardasi and Gornick, 2000; Gornick, 1999), and that an important component of the growth in part-time work is the growth of the service sector in the 1980s (Smith, Fagan and Rubery, 1998). We can see from figure 7 that women are much more likely to be working part-time than are men (Gornick, 1999). Under 10 per cent of men work in part time jobs, and, although the proportion has increased very slightly from 6.9% to 9.4% in the last ten years, the proportion of women in part time-work (44 per cent), however, is considerably higher.

Figure 7: Men and Women in full time and part time employment 100.0 90.0 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 % 0.0 1992 1993 1994 1995 1996 1997 1998 1999 2000 Year 2001 2002 Men:In full time employment Women:In full time employment Men:In part time employment Women:In part time employment Source: LFS Spring quarters March- May Seasonally adjusted An overriding reason for this is because women maintain primary responsibility for childrearing: researchers have demonstrated that most women in Britain work part- time at some point in their life, usually during the period when the child is very young; childbearing often involves a transition from a full-time to a part-time job (Blackwell, 2001; Joshi, 2002). For men, however, the situation is very different. Table 4 demonstrates that mothers are more likely than all working age women to work part time. However, fathers with young children are less likely than all working age men to work part time, but substantially more likely when the youngest child is older; a quarter of men with their youngest child aged 16-18 in full time education work part time. Table 4: Proportion of people in part time employment by age of youngest child, UK Women Men ALL All people of 42 9 24 working age (% in employment) Under 2 years 63 5 26 2-4 years 68 5 31 5-9 years 63 6 34 10-15 years 54 11 32 16-18 years in full 55 24 39 time education Base: All people of working age (men 16-64, women 16-59) in employment Source: Estimates form the Labour Force Survey (Autumn, 2002)

Figure 8 demonstrates clearly the pattern of mother s labour market participation by age of the youngest child. We can see that the proportion working full time gradually increases with the age of the youngest child as more time becomes available for the mother to move into the labour market, but the proportion working full time is highest for women with no child caring commitment. The proportion working part time peaks when the child is aged 5-10; this is because many women still take a complete break from the labour market when the child is very young and return to work initially on a part-time basis when the child enters school, so that they can still combine the caring role with work. The proportion of mothers working part time then gradually decreases, as the child gets older and the need for care diminishes, facilitating a move from parttime work into a full time work. Figure 8: Women 1 in part time, full time employment and economically inactive but looking after the family and age of youngest dependent child, Spring 2001 United Kingdom. No dependent children 16-18 11-15 5-10 Under 5 0 10 20 30 40 50 60 % Working full time Working part time Looking af ter family /home 1 Aged 16 to 59. Source: Labour Force Survey, Office for National Statistics The pattern of part time work, however, diffe rs according to marital status. As shown in figure 9, married/cohabiting mothers are more likely than single mothers to be working part time, regardless of the age of the youngest child. One reason for this is perhaps beca use they have partners who w ork, and therefore, unlike lone mothers, do not need to rely on their own earnings for a h ousehold income. Lone mothers, on the other hand, are less likely than married/cohabiting mothers to be working in either part time or full time work when the child is very young. The possible reasons for this are discussed in section 4 below on workless household.

Figure 9: Part time and full time employment of women 1 by marital status and age of youngest dependent child 2, Spring 2000 United Kingdom. No dependent children 16-18 11-15 5-10 Under 5 0 10 20 30 40 50 60 % Not married/cohabiting: w orking full time Not married/cohabiting: w orking part time Married/cohabiting: w orking full time Married/cohabiting:w orking part time 1 2 3 Aged 16 to 59. including 16-18 year olds in full-time education. separated or divorced. Source: Labour Force Survey, Office for National Statistics Includes single, widowed, Hours worked The Labour Force Survey, the source from which the above statistics were obtained, uses a self-defined notion of part time and full time work. If someone is categorised as working part time or full time it is because they state that they are doing so. This therefore gives little indication of actual hours worked. However it is important to explore the number of hours worked, rather than merely employment status, because, in reality, the number of hours that part timers and full timers work varies between and within each category. Moreover, the number of hours worked can influence earnings and, in the UK, a 16 hours working rule applies to benefit eligibility; people working less than 16 are able to apply for social assistance. The number of hours in paid work can also influence the balance of time spent with the family, and especially with children: working very long hours restricts the time available to spend with the family whilst working very few hours frees up more time to spend doing unpaid work and childcare in the home. However, those women who work part time on few hours are more marginalised in the work organisation that those who work part time on more hours or have full time jobs, and they have greater difficulties in being integrated fully in the labour market at a later stage (Rubery et al, 1998). We can see from figure 10, that the actual average weekly hours worked is over ten hours more for men than for women, and this figure has remained relatively constant throughout the last ten years.

Figure 10: Actual average weekly hours worked, 2000 45 40 35 30 25 Hours 20 15 10 5 Males aged 16-64 Females 16-59 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 year Source: LFS Spring quarters March- May Seasonally adjusted Figures 11 and 12 give a more comprehensive picture of the usual weekly hours worked by men and women over the last decade. It is immediately apparent that, although the highest proportion of both men and women work 31-45 hours a week, the picture is very different between the sexes for the remaining population. Whilst women s part-time work is characterised by very low hours, thereby reducing the overall average weekly hours of work for women, men's full-time work is characterised by very long hours. In 2002, a third of men worked more than 45 hours a week compared to 10 per cent of women. However, 42 per cent of women were working less than 30 hours - 13 percent less than 16 hours compared to only 9 per cent of men. This picture has not varied greatly in the last ten years, although the proportion of men working very long hours has, since 1997, begun to gradually decrease since 1997, and the number of women working very low hours (under 16 hours) has ever so slightly decreased. Nevertheless, the dramatic contrast between men and women s working hours remains.

Figure 11: Usual weekly hours worked: Men, 2000 70 60 50 40 30 20 10 % 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year less than 6 hours 6-15 hours 16-30 hours 31-45 hours over 45 hours Source: LFS Spring quarters March- May Seasonally adjusted Figure 12: Usual weekly hours worked: Women, 2000 60 50 40 30 20 10 % 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year less than 6 hours 6-15 hours 16-30 hours 31-45 hours over 45 hours Source: LFS Spring quarters March- May Seasonally adjusted The number of hours worked varies according whether a women has children, the age of the child and her marital status. Table 5 shows that married/cohabiting mothers are more likely than lone mothers to be working under 16 hours. Whilst lone

mothers with younger children are most likely to be working between 16-29 hours, married/cohabiting mothers are more likely to be working over 30 hours. Table 5: Number of hours worked by marital status of mother and age of youngest dependent child (2000) Lone mothers Married/cohabiting mothers Age of youngest dependent 19 years or 5 years or 19 years or 5 years or child younger younger younger younger Employed 50 32 70 60 Employed under 16 hours 15 16 20 23 Employed 16-29 hours 38 47 36 37 Employed 30+ hours? 48 38 45 40 Source: LFS Spring 2000 Reasons for Part-time working Part-time work has many disadvantages attached to it, perhaps the most crucial being that such jobs tend to be in low paying, low status, feminised occupations (Blackburn et al, 2001). A shift from full-time to part-time work after childbearing often involves a downward shift to a lower status occupation (Blackwell, 2001). However, Hakim (1996) argues that most part-time workers voluntarily choose part-time work: despite the disadvantages, part time work is a convenient way for mothers to reconcile work and childcare. Moreover, she claims that female part-time workers have different tastes, commitments and preferences about work compared to female full-time workers, who are more like their male counterparts. The Labour Force Survey asks why people work part-time although there are recognised problems with this question in that it provides too superficial an analysis of the possible reasons. With this caveat in mind, it shows that the majority of people say they work part-time because it is their preferred option; they didn t want a full time job. It can be seen from figures 13 and 14 that this statement is made by the vast majority of women working part time. Men, however, are more likely to give a variety of reasons, although the proportion stating that they couldn t find a full time job has declined from 30 per cent in 1993 to 16 per cent in 2002 (figure 14).

Figure 13: Reasons for Part-time working: women 90.0 80.0 70.0 Women: couldn't find full time 60.0 50.0 Women: didn't w ant full time 40.0 30.0 Women: ill or disabled 20.0 10.0 Women: student or at school % 0.0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 year Source: LFS Spring quarters March- May Seasonally adjusted Figure 14: Reasons for Part-time working: Men 50.0 45.0 40.0 Men:couldn't find full time 35.0 30.0 Men:didn't w ant full time 25.0 20.0 15.0 Men: ill or disabled 10.0 5.0 % 0.0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 year Men: student or at school Source: LFS Spring quarters March- May Seasonally adjusted

Other researchers (Bardasi and Gornick, 2000) question what voluntary really means. Burchell, Dale, and Joshi (1997) observe that while labour force surveys attempt to distinguish between voluntary and involuntary part-time work, they do not ask why respondents might state a preference for part-time work - that is, whether this was a forced choice or their own preference (ibid., p.217). It is recognized that preferences are expressed from the vantage point of current circumstances, needs and the range of perceived alternatives, and therefore would be expected to change in difference circumstances (F agan, 2001). Therefore, it could be argued that, rather than working part-time voluntarily, some mothers may choose to work part-time because there is no realistic alternative. In addition, researchers argue that work-time attitudes have to be interpreted in the context of society and at an individual level. For example, at the individual level both men and women s employment commitments are influenced by occupational position and domestic circumstances (Fagan, 2001). It is therefore misleading to use employment status (whether someone works full or part-time) and gender as an explanation for their commitment to work. At societal level, differences are observed in relation to preferences, part-time work patterns and child rearing across countries, which questions the homogeneity of part-time women s working time preferences and suggests that social policies, gender norms, and working time regimes are all influential (Fagan, 2001). In the UK, a lack of adequate and affordable childcare provision (Cohen, 1993) prevents feasible alternatives to part-time work. To illustrate this, Burchell, Dale, and Joshi (1997) point to a UK survey that found that 14 per cent of women part-timers would like full-time work but were prevented from seeking it by domestic commitments (p. 217). Moreover, family considerations also affect men s evaluation of the conveniences of work schedules, although women are more likely to be influenced by childcare and domestic responsibilities (Fagan, 2001). Full-time working hours are very long in Britain compared to other European Union countries and part-time hours are very short. Empirical research has found that, when questions about working hours are asked, both male and female part-timers (especially those in low paid manual jobs) invariably want to work longer hours, whilst both male and female full-timers (especially those in well paid managerial and professional posts) want their hours shortened. This not only questions the assumed gender differences in working patterns but also highlights that the choice for mothers is to undertake either very long or very short hours when neither alternative is actually the preferred option (Fagan, 2001). Temporary employment In the Labour Force Survey, temporary working is an umbrella title for fixed period contract work, casual work, agency temping, seasonal work and other temporary working, which can be very different kinds of work, and are taken up by different groups of people. In 2000, 7 per cent of the work force sampled considered their job to be temporary. We can see from figure 15 that in the last ten years, fixed period contract work has made up the highest proportion of temporary work (about a half). This kind of work peaked in 1995 at 54 per cent of all temporary work, but declined to 48 per cent by 2001. An increasing proportion of temporary employment is agency work, which has seen a ten percentage point increase from 7 per cent to 17 per cent

between 1992 and 2001. Workers in the public sector are more likely to be on fixed term contracts than those in the private sector (Bradshaw and Davis, 2002). Figure 15: Type of temporary employment 60 50 % of all temporary 40 30 20 10 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Years Fixed period contract Casual w ork Agency temping Seasonal Other temporary w ork Source: LFS Spring quarters March- May Seasonally adjusted The level of temporary work is relatively similar between men and women aged 16-64. In 2002, 8 per cent of women and 6 per cent of men undertook temporary work. However, the reasons that they undertaken temporary work are slightly different, as shown in figure 16 and 17. At the beginning of the 1990s, the reason that men gave for undertaking temporary employment was that they could not find a permanent job, and by 1995 half of all men in temporary work stated this. However, by 2001, this had declined to 32 per cent. The proportion of females saying that they could not find a permanent job has also declined, from 37 per cent in the early 90s to 23 percent. But the proportion of men saying that they were undertaking temporary work because they did not want a permanent job has increased from 19 per cent in 1992 to 26 per cent in 2002, whilst the proportion of women stating this remained relatively steady, at about a third.

Figure 16: Reasons for temporary employment: Male 60.0 temporary employees % of all male 50.0 40.0 30.0 20.0 10.0 0.0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year did not find permanent job did not w ant permanent job had contract w ith period of training other reason Source: LFS Spring quarters March- May Seasonally adjusted Figure 17: Reasons for temporary employment: Female 45.0 40.0 % of all female temporary employees 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year did not find permanent job did not w ant permanent job had contract w ith period of training other reason Source: LFS Spring quarters March- May Seasonally adjusted Other atypical and flexible working patterns In addition to part time and temporary work, other atypical and flexible working patterns included home-working, shift work, staggered hours, tele-working,

compressed hours, flexitime, job-sharing, term time working and annualised working. We discuss some of these below. Homeworking covers those who work mainly in their own home but excludes those who work in the same grounds or buildings as their home, those who work in different places using their home as a base and those who sometimes do work at home. Homeworking enables people to combine work with family life. Overall, women are more likely to be working at home than men, although the gap closed slightly between 1996 and 2002. This varies by occupation: women in managerial; professional; craft and related; personal and protective and sales occupations are more likely than men to be homeworking. Table 7: Homeworking 1 : by gender and occupation, 1996 and 2002 (Spring) 1996 2000 Males Females All Males Females All Managers and administrators 1 4 3 2 4 2 Professional 2 7 4 2 5 3 Associate professional and technical 2 2 2 3 2 3 Clerical and secretarial 4 5 5 5 4 4 Craft and related 1 5 4.. 5 4 Personal and protective.. 8 1 0 8 1 Sales.. 5 3.. 4 3 Plant and machine operatives 2.. 1.. 1 1 All occupations 2 1 4 3 2 4 2 1 Percentage of those in employment in each occupation who were homeworkers excluding those on government training and employment schemes. 2 Includes those with other occupations and those who did not state their occupation Compressed hours allow individuals to work their total number of agreed hours over a shorter period. For example, employees might work their full weekly hours over four rather than five days. They would be paid for a full-time job but would not receive overtime payments for the extra hours they work in any one day. Flexitime gives employees choice about their actual working hours, usually outside certain agreed core times. Individuals are paid for the hours that they work. Job-sharing typically involves two people employed on a part-time basis, but working together to cover a full-time post. Both receive pay for the hours they work. Term-time working allows employees to take unpaid leave of absence during the school holidays. Annualised hours describes working time organised on the basis of the number of hours to be worked over a year rather than a week; it is usually used to fit in with peaks and troughs of work. Pay will depend on the hours worked each pay period. Figures 18 and 19 show the flexible working patterns by gender for full time and part time workers. For both full time and part time workers, women are more likely than men to undertake flexible working patterns; 17 percent of full time men compared to 26 per cent of full time women and 15 percent of part time men compared to 25 per cent of part time women. Women working full time are especially more likely to be undertaking flexi-time and term time work and women working part time, term time and job-sharing.

We also know that flexible working is more common among employed parents than employees without dependent children: sixty per cent of employees with dependent children under 19 have a flexible working pattern compared with 42 per cent of employees without dependent children. The number of children also has an impact upon flexible working practices: employed parents with two or more children are more likely to have a flexible working pattern than any other group. Also, women in the public sector (47 per cent) are far more likely than those in the private sector (16 per cent) to work a flexible pattern, particularly term- time working, but also flexitime, annualised hours and job sharing. Figure 18: Employees with flexible working patterns: by gender, Spring 2001 Full time Employees Nine day fortnight Term-time w orking All employees Females Males Four and a half day w eek Annualised w orking hours Flexible w orking hours Any flexible w orking pattern 0.0 5.0 10.0 15.0 20.0 25.0 30.0 Source: Labour Force Survey, Office for National Statistics %

Figure 19: Employees with flexible working patterns: by gender, Spring 2001 Part time Employees Job sharing Term-time w orking All employees Females Males Annualised w orking hours Flexible w orking hours Any flexible w orking pattern 0.0 5.0 10. 0 15.0 20.0 25.0 30.0 source: Labour Force Survey, Office for National Statistics Latent demand for part time and flexible working patterns There remains, however, a considerable unmet demand for flexible working. Research conducted by the Government indicates that the demand is hig h for both men and women and covers a wide range of working patterns. Table 5 shows that men are mostly interested in compressed working weeks and flexitime and women in flexitime and part-time working. The latent demand for flexible working, both for parents and those without children, is not great, with the exception of part time and job-share working which are more popular among parents. Table 8: Latent Demand for Flexible Working Working Practice Males Females All Annualised Hours 24 18 21 Compressed Week 40 30 35 Job-Share 13 20 16 Reduced Hours 23 26 24 Term-Time 22 28 25 Flexitime 48 44 47 Part-time 21 35 26 Source: Labour Force Survey, Spring 2001 Note: Percentage wanting to work this practice. 4 Non-employment/ workless households Women s, and particularly mothers, integration into the labour markets has consequences at the household level. There are now more people in employment than at any other time in the post-war period, and most working age households in the United Kingdom are work-rich - that is, where at least one person is of working %

age and all people of working age are in employment. Work-rich households, as a proportion of all working age households, has risen from 50 per cent in spring 1992 to 58 per cent in spring 2002. The proportion of working age people living in households with no one in employment has not changed in the past ten years. at around 16 per cent In spring 2002. However, among lone parent households with dependent children, the proportion of workless is much higher, at 44 per cent in spring 2002, although this is 10 percentage points lower than in 1992. The proportion of children in workless households has been falling since the mid-1990s and remained stable between 2000 and 2002. The changes in all these proportions can be attributed partly to changes in economic activity and partly to changes in household size and structure over time. However, the reason why such a high proportion of lone parents are workless warrants special attention. Research has shown that they are more likely to be lack marketable qualifications (Finch et al., 1999; Lewis et al., 2000; Dawson et al., 2000; Holtermann et al., 1999); to live in metropolitan areas (Holterman et al., 1999), which experience lack of labour demand (Turok and Edge, 1999); to be in ill health (Lewis et al., 2000; Dawson et al., 2000), or to have a child or someone else in the house whose illness restricts opportunities to work (Marsh, 2001); are more likely to experience low morale, which can be a barrier to work (Finlayson et al., 2000; Marsh et al., 2001). Perhaps another factor is that lone mothers, unlike married/cohabiting mothers, have the opportunity of caring for their children full-time and still receiving social security benefits until the child is 16 years old. A major barrier to work for lone mothers in the UK is, however, considered to be lack of affordable childcare (Holtermann and Clark, 1993). Marsh (2001) identified that a third of out-of-work lone parents cite a lack of affordable childcare as a barrier to work. However, only a minority of these said it was the sole barrier and lone parents long established in work rarely cite childcare as a major difficulty that they had to overcome to enter and/or to keep paid work. Rather, choice contributes to the decision not to work or to delay work. The opportunity to spend as much of their own time with their children as possible is important despite the financial advantages work offers (Marsh, 2001). Lone parents do not look for childcare to fit in with their prospective employment, rather they look for jobs with hours that enable them to look after their children themselves (Finlayson et al, 2000). It seems that many lone mothers want to look after their children themselves and that childcare only becomes an issue once work is considered, not before.

Figure 20: Working age households: 1,2 by household economic status (United Kingdom) 70 60 50 40 30 20 10 % 0 1992 1997 2000 2001 2002 year Households w ith all in employment Workless households Working-age people living in w orkless households Children in w orkless households Workless lone parent households w ith dependent children 1 Percentages have been adjusted to include estimates for households with unknown economic activity. Percentages based on all working age households, at spring each year. 2 These estimates are not seasonally adjusted and have not been adjusted to take account of the recent Census 2001 results. See Appendix. Part 4: LFS reweighting. Children in workless households is as a percentage of working age households. Workless lone parent households is as a percentage of all one parent households. Source: Labour Force Survey, Office for National Statistics Figure 21 shows the proportion of couples with dependent children with different number of earners in the household. We can see that, as a result of the growing proportion of mothers entering the labour market, the majority of couple families are dual-earner families, and the proportion has gradually increased over the 1990s. Indeed, this increase has been in expense of the male breadwinner family. Nevertheless, a quarter of UK families with dependent children still uphold the male breadwinner family, with the woman not working. This suggests that, in these families, it is the women who are staying at home to look after the family. However, the male is not earnin g in only a small percentage of UK families. The proportion of households with neither parent in work has remained relatively steady, at around 10 per cent, although with a slight increase in the mid 90s.

Figure 21: Couples 1 with dependent children by number of earners in the household (United Kingdom) 70 60 children % of couples with dependent 50 40 30 20 10 0 1988-89 1989-90 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1998-99 Year Tw o earners No earners Male only earners Female only earners 1 Married and cohabiting couples with men aged 16 to 64 and women aged 16 to 59. Data not available for 1997-98. Source: General Household Survey, Office for National Statistics 5 Maternal/paternal and parental leave A major reason for the increase in women s labour market participation is that more women are resuming employment after maternity leave. In 1996, 67 percent of mothers returned to work within 10/11 months of giving birth. But maternity and parental leave are included in employment figures and therefore need to be accounted for when analysing employment figures. New rules for maternity and paternity leave were introduced in April 2003, which are described in the policy section. The figures in this section are based upon the old rules, which are explained under each separate heading. The Work-Life Balance Survey asked employees about the type of leave they had taken in the last 12 months in 2000. Men and women were equally likely to have taken bereavement leave, but women were far more likely than men to have taken time off to look after children (figure 22). Figure 22: Proportion of employees aged 15 to 65 who had taken selected types of leave in the last 12 months by gender, Great Britain 2000

Career breaks Females Males Leave to look after others Type of leave Time off to look after children Bereavement leave 0 2 4 6 8 10 12 14 16 18 20 % Source: Work-life Balance [Employee] Survey 2000, Department for Education and Employment Maternity leave Under the old rules, statutory paid maternity leave could be taken for 18 weeks wit h the option to extend the leave for additional 22 weeks, whic h are unpaid. The length of maternity leave taken varies, and for many mothers the time they a re able to take is limited by financial constraints. Figure 22 shows that 51 percent of m others working at least 8 hours (93 per cent) had taken only the statutory paid leave of 18 weeks, or less. Sixty-one percent of mothers said that longer maternity leave was available to them but 78 percent of these said that they had not taken this up because of financial constraints. Older mothers, public sector workers, those in managerial and professional occupations and those in high income household were all more likely to take over 18 weeks leave. This is not surprising since public sec tor workers and those in managerial and professional occupations were more likely to receive extra financial assistance from their employer, and we can presume those in high income households were in a better financial position to take unpaid leave.

Figure 23: Length of maternity leave Great Britain Autumn 2000 All mothers working 8 hours per week who had taken maternity leave in previous 5 years n=183 less than 14 w eeks 14 w eeks 15-17 w eeks 18 w eeks 19-29 w eeks 30-40 w eeks 40+w eeks Source DTi Survey of how parents in employment balance work, family and home Paternity leave No statutory paternity leave existed before 2003, but employers could allow fathers to take discretionary paternity leave. According to a Department of Trade and Industry (DTI) study of working parents in Autumn 2000, 85 per cent of those working fathers able to take time off work when their last child was born (after October 1995) had taken some paternity leave. The average length of time taken was 7 days. Parental leave A new right to parental leave was introduced in 1999. This enables fathers and mothers of children born on or after 15 December 1999 to take up to 13 weeks unpaid leave before the child's fifth birthday, provided they have at least one year's service and subject to certain conditions about when the leave can be taken. Since statutory entitlement to parental leave is a relatively recent piece of legislation, information on its effect to date is limited. In particular, it is difficult to isolate the effect of a statutory entitlement to parental leave for new parents from pre-existing arrangement for paid or unpaid leave offered on a voluntary basis by employees in 1998, 28 percent of employees in work places with more than 25 employees said they had access to this arrangement (Cully et al, 1999, table 7.3). In 2000, after the statutory leave had been introduced, only 29 percent said that their employer provided parental leave. Only 12 percent of these mothers and fathers had taken to parental leave since the statutory leave had been introduced (9 months previous being asked). This equates to 3 percent of the working parents in the survey. Leave for sick children Statutory entitlement to (unpaid) time off to cope with family emergencies was introduced in December 1999, which includes leave to look after sick dependent

children (under 16 or 16-18 in full time education or training). The actual number of days leave is discretionary; up to a maximum of 10 days unpaid leave per parent per year. In 2000, 63 percent of working parents said their employers did provide time off for family emergencies and managers and professional were more likely than other occupations to say their employers did provide time off. Whereas 50 percent of working fathers said they were entitled to paid time off in addition to their annual leave entitlement, only 38 percent of mothers reported this entitlement, although mothers were more likely to be entitled unpaid time off (38 percent compared o 23 percent for men). This may be in part due to mothers being more likely to work part time. The DTI survey of working parents also provides an overview of how often parents face families emergencies that might require time off work. In autumn 2000, 49 percent of working parents said they had experienced a family emergency involving their children during the working week, such as a child being sick or childcare falling through. More mothers (54 percent) than fathers (45 percent) recall an emergency. Definitions of what constitutes an emergency may not be uniform. Of those taking time off in the last year, 75 percent of parents had taken time off to look after a child who had been ill, injured or assaulted. They had, on average taken 2.4 days off, although 46 percent had taken no more than one day off work. Just over half of fathers compared to 40 percent of mothers took only one day or less off and the few parents who had taken more than a week off work were almost exclusively women. 6 Income and earnings For women more than men, labour market participation does not guarantee an adequate income. This is demonstrated by the gender wage gap. The difference between women and men with regard to earnings, and specifically hourly earnings is a key aspect of the gender pay gap debate. We focus on differences in hourly earnings, which are central to any analysis of women s access to financial resources. The gender wage gap is therefore the difference between average female and male hourly earnings. The main reason for focusing on hourly earnings is that women and men have very different patterns of working time (see the discussion above). Thus women have much shorter average weekly working hours than men, with much larger proportions of women than men working part-time. Comparisons of weekly hours, therefore, reflect differences in working time patterns as well as the gender pay gap itself (Dench et al, 2002). Indeed, related to the above is whether the gender pay gap should be calculated on the basis of full-time earnings, part-time earnings or both. Anderson et al.(2001) point out that combining full an part time hourly earnings will be affected by a women s greater tendency to work part time than men and the lower hourly wages that part time work attracts. However, comparisons of the full-time earnings of women and men gives an incomplete picture of women s disadvantage in hourly earnings, given that a higher proportion of women than men are found in part-time work, with lower mean hourly rates. A direct comparison of hourly earnings between men and women in part-time jobs is difficult precisely because there are relatively low numbers of men working part-time hours. For this reason, mean hourly earnings of women in part-time jobs is often compared with mean hourly earnings of all men or men in full-time jobs. However, doing this will inevitably represent the combined