Warren: Women in Low Status Part-Time Jobs

Tracey Warren (2000) 'Women in Low Status Part-Time Jobs: A Class and Gender Analysis '
Sociological Research Online, vol. 4, no. 4, <http://www.socresonline.org.uk/4/4/warren.html>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 1/6/1999 Accepted: 28/10/1999 Published: 29/2/2000

Abstract

Why, given all the problems associated with part-time employment in Britain, do women work part-time at all? Does the answer to this question lie in gender-based explanations which focus on womenís caring responsibilities? This paper addresses these issues by focusing on the relative experiences of the largest group of part-timers, women working in low status occupations. It is concluded that a gender-informed analysis of womenís part-time employment is clearly vital, but an awareness of further dimensions of social inequality is required if we are to understand diversity amongst part-timers. Relative to full-timers, part-timers are similar in their life-cycle positions, their marital status and motherhood status. However, incorporating a class analysis shows that part-timers in lower status jobs stand apart in that they are disproportionately likely to have been brought up in working class households and, as adults, they are more likely to be living in very low waged households with partners who are also in low paid manual occupations. It is concluded that women go into the lowest status part-time jobs in specific social contexts and, as a result, we cannot lump together into one unified group, women working part-time in manual and higher status occupations, and then talk sensibly about part-time work and its impact on women. It is essential to examine the interaction of gender and class inequalities to better understand these womenís working lives.

Keywords:

Class; Diversity; Gender; Part-Time Employment

Women in Low Status Part-Time Jobs: A Gender and Class Analysis

1.1

Why, given all the problems associated with part-time employment, do women work part-time at all? This remains a major question in research into women's employment despite increased awareness of the existence of heterogeneity in part-time work, both in terms of the quality of jobs labelled as part-time and in the types of women who enter these jobs. Gender-based answers have been dominant. One argument has been that female part-timers are particular types of women who prefer to work part-time as it means that they can prioritise their homes and still earn a small wage. An opposing argument has been that many women with family responsibilities have no choice, they are forced into jobs of short hours. This paper is located within these debates but it is also stimulated by an interest in diversity amongst part-timers. It argues that dominant gender based analyses are vital to develop an understanding of why women work the hours they do. However crucially, in order to understand diversity amongst part-timers and, in particular, why some are in low status jobs whereas others enter higher paid and higher status occupations, a new factor comes into play in addition to gender: women's class.

1.2

The study of women's part-time employment saw a major boost of activity with the publication in 1991 of the first of Catherine Hakim's series of 'grateful slave' and 'preference theory' papers, and her suggestion that part-timers and full-timers are qualitatively different types of women with different attitudes to working and home life. A number of studies resulted that disputed Hakim's suggestion, with the majority focusing on women in high status jobs. These papers concluded that not only can part-timers be as work committed as full-timers, but that part-timers are a diverse group of women. They display a variety of preferences and a range of attitudes to work (Caven 1998; Crompton and Harris 1998a; Walsh 1999). However, this move in sociology towards exploring variety in women's motivations for and experiences of part-time working has revealed relatively little interest in examining class divisions amongst part-timers. This neglect has occurred despite the fact that, whilst many recent studies have focused on high status jobs, the bulk of part-time jobs are low paid and at the bottom of the occupational hierarchy.

1.3

This paper aims to explore class diversity in the female labour market by first asking what conditions women face when they work in the lowest status part-time jobs. Given the disadvantaged character of these jobs, the next question addressed is why do women work here? Do the dominant gender-based explanations which focus on women's domestic responsibilities provide the answer? For example, it is known that women workers who have children are highly likely to work part-time to enable them to balance their home and working lives. So are female part-timers in the lowest status jobs in more intense motherhood stages than other part-timers? Furthermore, are part-time manual women able to take such poorly paying jobs because a high-waged male breadwinner supplements their own low wages? Or, conversely, are explanations other than those based on women's caring roles required to account for this diversity? Examining the relative experiences of the lowest status of women workers, the key issue of class divisions amongst female part-time employees is explored.

1.4

These research questions are addressed by analysing data on female part-timers and full-timers in the 1995 British Household Panel Survey (BHPS). The BHPS contains a number of key variables for the analysis of women's employment including occupation, wages and hours of work, as well as demographic and household characteristics such as education, marital status and the labour force activity of women's partners. By comparing manual part-timers with higher status part-timers, and also examining full-timers in different status jobs, it is possible to show the severe disadvantage that women in low status part-time occupations face in the labour market. It is also possible to question whether this disadvantage is related to women's caring roles by examining, for example, the number and ages of any dependent children. The paper rejects the idea that purely gender based explanations can explain why women are in the lowest status part-time jobs as part-time manual women were similar to other part-timers in their caring roles. However, part-timers differed in a number of class-related ways. First, the manual workers were more likely to be living in households where their partners were also in low paid manual jobs. Second, the women's severely disadvantaged labour market positions were linked to their having low levels of education. This low educational attainment was itself related to the fact that women in the lowest status part-time jobs were more likely to have been brought up in working class households than the other part-timers. It is concluded, therefore, that although part-timers were united in the impact of gender on their working lives, there was a large degree of class diversity amongst the employees. It is essential to examine the interaction of gender and class inequalities to better understand women's working lives.

Explaining Female Part-Time Employment.

2.1

Any explanation of the pattern of female employment in Britain needs to take into account women's supply-side characteristics and their gender roles when living under such a strong male-breadwinner regime (Lewis 1992). However, only focusing on supply issues can result in an over-concentration on how choices and preferences shape women's working lives. Therefore, to understand how and why women work part-time, it is also necessary to explore the ways in which supply factors interact with demand-side factors - including whether and in what form employers offer part-time jobs (see Fagan and O'Reilly 1998 for a review of supply and demand debates surrounding part-time employment). A stress on the importance of supply factors was popular in the post-war years of British sociology, for example in the 'women's two roles' and 'human capital' theories^[1], but it has re-emerged in the 1990's in the well-known work of Hakim (1991, 1995, 1996a, 1996b, 1997, 1998a, 1998b).

2.2

Stimulated by the existence of differences amongst female employees, Hakim devised typologies of women workers. She has argued that there are qualitatively different groups of women in the labour market with different desires and this is leading to a polarisation of female employment in the 1990's. Highlighting that many female part-timers seem satisfied with inferior jobs whilst full-timers are more likely to be in long-term, male-like 'continuous career' routes, Hakim has argued that most full-time women have similar 'work commitment' or 'conventional career orientations' to men. This, she states, explains why they are often in higher status and better paid jobs. Conversely, women in half-time (less than 30 hours a week) or marginal (up to 10 hours a week) part-time jobs are more likely to be in 'modern home-maker careers' in which they combine family with work but prioritise the former. They have a 'purely instrumental' approach to work.

2.3

Although Hakim's thesis fits in with research into heterogeneity amongst women and the polarisation of women's experiences (see Glover and Arber 1995; Rubery et al. 1994 for example), she has been questioned. Indeed, a debate has resulted over the importance she attaches to choices and preferences in shaping women's entry into part-time or full-time jobs. Perhaps most notably, Ginn et al. (1996) and Crompton (1997) argue that the 'grateful slave' or 'modern home-maker' ideas of Hakim over-play women's orientations to work as one of the main influences on their employment experiences. They also under-play the idea that women's work choices are shaped by gender divisions of labour in the home and the labour market (see also Crompton and Harris 1998a, 1998b; Crompton and Le Feuvre 1996; Proctor and Padfield 1999; Scheibl 1998; Warren and Walters 1998).

2.4

Therefore, gender inequalities have been central in this 'grateful slave' debate. However, class inequalities and divisions amongst women have not, and this relative inattention to class is an important omission. For example, behind Hakim's creation of different typologies of women workers is the argument that certain women are not committed to employment: they know that they want a 'home-maker career' and so have little interest in building up their human capital. As a result, they have not been committed to the labour market and have not pursued formal qualifications, preferring to enter less taxing and lower status part-time occupations. Lying within this argument seems to be the belief that if women have commitment to getting on in the labour market then they will ensure that they receive a good education, attain good grades and get a good job. 'Aiming for a good job is a necessary first step towards getting one' (Hakim 1991: 114). But taking this to its logical conclusion implies that women have equal chances to get on in the labour market (and in education) should they wish to do so. For many sociologists studying inequality in Britain, whether along lines of class, gender, ethnicity or 'race', this argument is surely anathema^[2]. For example, it would be highly questionable to argue that working class children leave school with lower qualifications than middle class children simply because they have decided they do not want or need the qualifications as they are not aiming for good (middle-class) jobs^[3]. Yet surely this is what, in its extreme form, Hakim has been arguing with respect to female part-timers. Stressing both equality of opportunity and that it is commitment that leads to success neglects the fact that women's education and employment chances are shaped by their gender, their class and their ethnicity and 'race' (see Bhopal 1998 on how 'race', ethnicity and gender impact on women's educational outcomes).

2.5

The relative omission of a class-based discussion in the 'grateful slave' debate is somewhat surprising given the growing interest in diversity amongst female part-time employees (see Walsh 1999 for example). The neglect of class has occurred despite the fact that, in the main, the empirically founded critiques of Hakim have been based mostly on studies of women in high status jobs. In these studies, the polarising of women based on their attitudes and work commitment has been questioned and the very focus on attitudes and commitment has been problematised.

2.6

First then, focusing on the issue of attitudes and choices, it has been argued that part-timers are not uncommitted to work nor are full-timers necessarily as committed as Hakim has suggested. Discussing women working as architects, for example, Caven (1998) argued that many part-timers are clearly committed to their employment despite their part-time hours. Walsh (1999) demonstrated the divergent preferences which were expressed by a large group of women working part-time in banking and finance in Australia. Devine (1994) studied full-timers in science and engineering, showing that women did not arrive in these high status jobs just because of their pure commitment to work, and their choices. Instead, particular social contexts, including middle-class upbringing and being educated at single-sex grammar schools, had allowed the women to exercise their choices.

2.7

A second conclusion of these studies is that agency, attitudes, preferences and work commitment are not the whole nor even the major part of the 'work part- or full-time' picture for women workers (see Dex 1988). Proctor and Padfield (1999) have demonstrated the necessity of examining the interplay between agency and structures in women's working lives. Crompton and colleagues have shown similarly that knowledge of work commitment does not explain why women in certain high status occupations are more able to work part-time when they have children than are other high status women. It is necessary to examine whether part-time jobs are made available for women too. For example, pharmacy as a profession is actually constructed to provide women with this type of flexible employment. However, part-time opportunities have not been created in higher status managerial banking occupations, and so the choice to work part-time is severely restricted here (Crompton and Harris 1998a; Crompton and Le Feuvre 1996).

2.8

Therefore, there have been useful conclusions on what factors shape women's entry into, and experiences of, different types of part-time and full-time employment. Yet class has not emerged as a pivotal issue. This is despite the fact that class is central in social mobility research which is concerned, similarly, with how certain types of workers enter good or bad jobs. Studies of social mobility have long shown that employment opportunities and the choices available to women and men are strongly affected by class, For example, children of middle class parents are more likely to enter higher status occupations than are children from working class backgrounds (Halsey et al. 1981; Marshall et al. 1988). This is because middle-class children are more likely to attain better educational qualifications but also because they enter higher status jobs than working-class children even controlling for their educational levels (Egerton 1997; Swift and Marshall 1997).

2.9

So, to understand why certain workers end up in higher or lower status jobs, it is important to examine their class background as well as their gender roles. Studying class divisions amongst women can only improve our understanding of how and why women enter the jobs that they do. This is particularly appropriate given that class divisions amongst the female labour force are becoming more extreme as certain groups of high status women are improving their positions in the labour market (Rubery and Tarling 1988. See also Crompton and Sanderson 1990; Witz 1992). Gender based explanations may constitute a major reason why women work part-time rather than full-time, but a class based analysis of this process is vital too because 'the experience of gender is class-specific' (Bradley 1996: 109). Indeed, Glover and Arber (1995) and MacEwan Scott and Burchell (1994) have demonstrated how women in senior occupational classes are more able to avoid part-time working because they are more able to minimise the effects of motherhood on their working lives. As Bradley (1996: 110) has stated, 'at the bottom of the class hierarchy, the weight of gendered inequality is disproportionately felt'.

2.10

In summary, there is a growing recognition of the need to examine diversity amongst female part-timers and to accept that part-timers are not uniformly different to full-timers in their experiences and attitudes (Warren and Walters 1998). However, the majority of British female part-timers are still concentrated in lower status jobs (Burchell et al. 1997; Martin and Roberts 1984; Fagan and Rubery 1996). Therefore, by studying specifically more 'typical' part-timers - women working part-time at the bottom end of the labour market - this paper questions what an analysis informed by awareness of gender andclass can add to our current understanding of who works part-time, why and with what consequences. Focusing on the relative positions of women in part-time manual jobs highlights how supply-heavy theories stressing women's choices neglect the fact that the choices which women can make are shaped by their class too, and not only by their family responsibilities. There is a large degree of class diversity amongst female employees, and it is not possible to understand women's varying work experiences without reference to these divisions. Therefore, the interaction between women's agency and social structures - structures of gender and class and not gender alone, mean that certain women enter into some of the most disadvantaged positions in the labour market^[4].

Data and Methods

3.1

The research is based on analysis of the British Household Panel Survey (BHPS) which is a large, nationally representative data-set suitable for cross-sectional and/or longitudinal analysis (see Taylor et al. 1991 for details on the sampling). Focusing on respondents who gave a full interview^[5] in the fifth wave of the BHPS (Wave E, 1995), a sub-sample of female employees of standard working age (between 18 and 59, excluding full-time students) was constructed. This sub-sample contained over three thousand women. Sixty-two per cent were in employment and had data available on their working conditions including their wages, hours of work, and on their household characteristics such as family circumstances and employment details of any partners.

3.2

The aim in the paper is to study women in different 'quality' part- and full-time jobs (where part-time has been defined as working usual weekly hours of thirty or less^[6]). Therefore, the sample has been divided into different status occupational categories based on the nine groups of the 'Standard Occupational Classification' variable (SOC. See Table 1). Continuing to use this variable throughout the analysis, although it would retain detail, would mean very large tables on cross-tabulation. Therefore SOC was recoded. As part-timers dominated most of the occupations at the bottom manual end of the occupational hierarchy (excluding 'Craft and related' and 'Plant and machine operatives', where numbers of women were small overall), the standard manual/non-manual divide was suggested. However, this approach is problematic as it groups together women in very diverse occupations such as professional and clerical workers. The wages of women in the different SOC groupings were examined as a proxy for job quality, and Groups 1-3 emerged with median hourly wages of Ł7.60 or more, compared with around Ł5.50 for clerical workers and a range of Ł3.50 to Ł4.50 for the five manual groups (there was less diversity within the manual occupations in Glover and Arber's, 1995 study of part-timers too).

Table 1: Occupational distribution of female employees aged 18 to 5

SOC		Part-time	Full-time	All
		Column %

	Higher Level Non-Manual
I	Managers and administrators	3 (14)	13 (86)	9
II	Professional occupations	9 (33)	12 (67)	11
III	Associate professional and technical	8 (30)	13 (70)	11

	Lower Level Non-Manual
IV	Clerical and secretarial	24 (34)	33 (66)	29
	Manual
V	Craft and related	2 (28)	3 (72)	2
VI	Personal and protective service	20 (56)	11 (44)	15
VII	Sales	18 (69)	6 (31)	11
VIII	Plant and machine operatives	3 (26)	6 (74)	5
IX	Other	15 (79)	3 (21)	8

	N	876 (42)	1221 (58)	2097

Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

3.3

Reflecting the above wage split, SOC was trichotomised into higher status non-manual, lower status non-manual and manual workers: a 30:30:40 distribution resulted. Combining the trichotomous occupational grouping with the part-time/full-time variable produced the six groups of women which are the focus of the analysis (see Box 1). Part-time manual work emerged repeatedly as the most disadvantaged employment category for women as a result of the combination of the disadvantage of working part-time hours and in low status jobs. Hence it is not just gender-based reasons which are needed to understand these women's labour market locations. It is necessary to incorporate gender-based explanations (shaping number of hours worked) and class (occupational status) to account for the severe disadvantage they face. This next section explores the extent and nature of this disadvantage.

Box 1: Occupational and working time groups analysed

Part-time

Managerial, professional/associate professional and technical
Clerical
Manual

Full-time

Managerial, professional/associate professional and technical
Clerical
Manual

The Relative Labour Market Locations of Women's Part-Time Jobs.

4.1

A great deal of previous research has demonstrated that, in general, British part-timers are in inferior positions in the labour market compared with full-timers (Burchell et al. 1997; Hakim 1997; McRae 1995; Rubery et al. 1998). This research reaffirms these findings. Women were collected into higher and lower status groups, but even within these categories part-timers were over-concentrated on the very lowest ranks. More of the part-timers than full-timers in manual jobs were in the 'other' occupation (cleaners, domestics, catering assistants, counterhands and shelf fillers) for example (Table 1 earlier). Proportionally fewer of the part-time managerial/professional workers were in the highest level 'managers and administrators' occupation^[7]. The over-concentration of low status part-timers at the very bottom of the occupational hierarchy and the severe disadvantage they faced resulted in a wide class diversity amongst the part-timers which will be seen throughout the paper.

4.2

Moving on to examine the working conditions of the six occupational/working time groups, the paper shows the intense degrees of labour market disadvantage faced by women working part-time in the lowest occupational categories. A wealth of indicators demonstrated their disadvantage. The ones presented here are women's typical working weeks and their wages - to indicate the inferior conditions of their jobs, and access to a company pension - to demonstrate the equally problematic longer term implications of this type of work. Beginning with women's typical working weeks, part-time manual women's employment schedules (as evidenced by their usual hours of work and the times of day worked), were the most 'non-standard' of the six groups of women studied. Working very few hours a week is well known to be associated with negative consequences for workers. In the past it meant that they faced exclusion from a great deal of employment and social protection including access to rights and benefits in Britain. Recent changes have outlawed most direct discrimination based on hours worked (Cook 1998; Heery 1998), but indirect discrimination remains. Therefore, the typical number of hours women work remains an important indicator of the quality of their employment. Those in jobs with non-standard hours tend to be over-concentrated in lower status employment (Dex 1987, 1988).

4.3

In Table 2, part-timers' and full-timers' hours have been disaggregated into short and longer hours bands and this reveals distinct class diversity amongst the women. Female part-timers in manual occupations were working substantially shorter average weeks than other part-timers, and shorter than professional part-timers in particular. Indeed, these two groups were concentrated at opposite poles of the part-time hours range. Two-fifths of the professional women were working long part-time hours whilst two-fifths of the manual women were working short hours. There were also distinct differences in hours worked by occupational class amongst the full-timers, although there was less evidence of polarisation here. Substantially more manual full-time women were working longer (at least 37) hours a week, no doubt to help them 'get by' on low hourly wages (the wages of the six groups of women will be discussed later).

Table 2: Hours bands of female employees aged 18 to 59, by occupational status

	Prof.	Clerical	Manual	All
	Column %

Part-time employees
Short part-time (0-15)	21	24	41	33
	(13)	(18)	(69)	(100)
Moderate part-time (16-23)	39	53	36	41
	(18)	(31)	(51)	(100)
Long part-time (24-30)	39	23	23	26
	(29)	(21)	(50)	(100)

Median hours	20	20	17	20
N	168	209	499	876
Total %	8	10	24	42

Full-time employees
Short full-time (31-36)	56	71	43	57
	(37)	(41)	(22)	(100)
Long full-time (37-44)	38	29	54	40
	(37)	(24)	(39)	(100)
Very long full-time (45+)	6	0	3	3
	(69)	(5)	(26)	(100)

Median hours	37	37	38	37
N	467	399	355	1221
Total %	22	19	17	58

Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

4.4

Part-time manual women were concentrated in jobs with the shortest hours and these, in general, are the most unpopular of part-time hours (see Horrell et al. 1994). Number of hours are not the only indicator of an 'atypical' week; there is also variation in the times of day worked. The six groups of women were examined to see who were more likely to work unsocial hours outside the 'standard' nine to five day, which Fagan (1999: 5) has argued conflict 'with the temporal rhythms of social life' and are associated with both health and domestic problems. In the BHPS, women were asked to describe the time of day they worked. They were grouped into those who worked during the day, those working in the evening or at night, and those whose times varied or were mixed such as a combination of evenings and lunchtimes (Table 3). The women least likely to have standard hours were the part-time manual workers (see Glover and Arber 1995 too), mostly due to higher proportions working evenings or at night, which would be common for cleaners for example. Indeed, the majority of women working evenings/nights were manual part-timers (61%).

Table 3: Times of day worked by female employees aged 18 to 59, by occupational status

			(Percentages)
	Part-time			Full-time			All

	Prof.	Clerical	Manual	Prof.	Clerical	Manual	Part-time	Full-time	Total
Day	75	85	66	80	95	73	72	82	78
	(7)	(11)	(20)	(23)	(23)	(16)	(39)	(61)	(100)
Evening/night	9	4	14	1	0	5	10	2	5
	(13)	(6)	(61)	(5)	(0)	(14)	(81)	(19)	(100)
Mixed/other	16	11	20	19	5	23	17	16	16
	(8)	(7)	(29)	(26)	(6)	(24)	(44)	(56)	(100)

N	168 (8)	209 (10)	499 (24)	467 (22)	399 (19)	355 (17)	876 (42)	1221 (58)	2097 (100)

Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

4.5

Perhaps the key indicator of the intense labour market disadvantage faced by female part-time manual workers is their very low wages. A main problem for low waged workers is that their gross weekly earnings lie below the Lower Earnings Limit (LEL) for National Insurance Contributions (NICs). If no NICs are paid then workers are not able to accumulate a state pension in their own right and, in addition, they are excluded from other benefits, for example State Maternity Pay, Statutory Sick Pay and Job-seekers Allowance (see Taylor 1998^[8]). Low weekly earnings are a result of a combination of short working weeks, discussed above, and low hourly wages. The wages of the women were examined, measured as usual wage or salary in current main job (pre-tax and deductions. See Appendix 1) and are analysed next.

4.6

Focusing on hourly wage rates as a comparable measure of diversity amongst the part-timers and full-timers, a distinct manual wage disadvantage united the women. As expected, women working in manual jobs (whether part-time or full-time) were earning substantially less than their clerical and managerial/professional counterparts (Table 4). Full-time manual women had some of the lowest hourly wage rates and so it is likely that part of the reason they were working longer weeks was to boost these earnings^[9]. Nevertheless, the lowest waged were the part-time^[10] manual women who faced double wage problems because of both the manual disadvantage and a part-time disadvantage. In fact, given their very low hourly wages and their short working weeks, almost half of the part-time manual women were earning weekly wages under the LEL (which was Ł58 in 1995), with all the problems of exclusion from rights and benefits discussed above. In contrast, a large majority of the other part-timers were comfortably over the LEL. Only 11% and 19% of the professional and clerical women respectively were earning less than Ł58. Due to the very low wages of the manual women in part-time jobs and full-time equivalent or even higher earnings of the professional women working part-time, wage diversity by class was more substantial amongst the part-timers than full-timers.

Table 4: Hourly gross wages of female part-time and full-time employees aged 18 to 59

	Part-time			Full-time			All

	Prof.	Clerical	Manual	Prof.	Clerical	Manual	Part-time	Full-time	Total

Median hourly gross wage (Ł)	8.7	4.9	3.6	8.4	5.6	3.9	4.2	5.9	5.1
Mean hourly gross wage (Ł)	9.4	5.1	3.4	9.0	5.8	4.5	5.2	6.7	6.1
Median wage as % of overall average (%)	171	96	71	165	110	76	82	117	100

N	168 (8)	209 (10)	499 (24)	467 (22)	399 (19)	355 (17)	876 (42)	1221 (58)	2097 (100)

Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

4.7

A further key example of the problems facing women in part-time manual work lay in access to company pensions. Despite the 1994 European Court of Justice ruling that employers cannot exclude part-timers, the right to a decent company pension is still governed by two other factors which disadvantage part-time workers and part-time manual workers in particular. First, a company needs to make a pension available to its employees. This 'perk' is strongly related to the status of the occupation and so workers in low status jobs fare badly. Second, earnings determine the level of (employer and employee) contributions to the pension scheme. The low pay associated with working part-time means that part-timers' contributions, and those of employers (who contribute a multiple of the employee rate) will be very low. As a result, most women who work part-time for a significant proportion of their working lives will accrue a meagre, if any, company pension in their own right by the time that they retire (Rowlingson et al. 1999 detail levels of pension savings for different life-cycle groups in the 1995/96 Family Resources Survey). Given the decline in the real value of the state pension, and the exclusion of part-timers earning below the LEL from state pensions, this remains a fundamental problem for women and for part-time manual workers in particular (see Ginn and Arber 1996, 1998).

4.8

In the sample, only two-fifths of part-time manual women were offered a pension by their employers and the variation amongst the part-timers was extreme on this measure - with a figure of fully 72% for professional workers compared with 62% for women overall. Most of the manual part-timers had no company pension and, in addition, many were earning under the LEL. Therefore, if they remain in these jobs long term then the majority will no doubt be expecting to depend on another pension source, most probably a male partner. However, it is unlikely that their male partners' pension incomes would be high as, it will be shown later, the men were low waged and in low status jobs too. So this examination of women's pension coverage has shown that unless part-time manual women increase their hours and/or enter higher status work, they seemed destined to carry their very low incomes into old age.

4.9

In summary, this section on the working conditions of the six groups of employees confirmed that women's experiences were divided intensely by their class. It demonstrated, in addition, that the most extreme class diversity lay amongst the part-timers. As a result, part-time manual work was by far the most disadvantaged employment category for women, more so than both other part-time jobs and manual full-time jobs. Therefore, the second main research question of this paper is why were these women working here?

The Women in Part-Time Jobs

5.1

Female part-time manual workers: female carers?
Were the poor jobs of women in manual part-time work related to their being concentrated in life-cycle stages where child-caring commitments were intense? Were they even more likely to be 'female carers' than other part-timers, with a male breadwinner and so, because of all these factors, they were busy prioritising other investments such as home skills? That is, were gender-based explanations enough to explain diversity amongst the women? The results do not support this interpretation as - in terms of a number of indicators of women's caring roles; their life-cycle stage, age, marital status, and the number and ages of any dependent children - similarities amongst the part-timers when compared with full-timers outweighed the differences between them (see the table in Appendix 2). Therefore, knowledge of women's gender roles as indicated by their life-cycle positions and child-care responsibilities can clearly contribute to our understanding of which women employees work part-time or full-time in Britain. However, these factors tell us little about diversity amongst part-timers and, in particular, why they work in very poor or in higher quality jobs.

5.2

Another major part of this 'female carer' question asks whether women work in such low paid jobs because they do not need the extra income, because they have a high-earning male breadwinner. The majority of female part-timers were living in a couple (Appendix 2), and the vast majority of these women's partners were employed (91%) and so dual-employee couples were focused upon. Men's monthly wage incomes were calculated and their occupations were examined using the same trichotomous classification used for women, higher and lower status non-manual and manual working. The class divisions which have been emerging amongst the part-timers throughout the analysis are readily apparent on these household issues too. Further, the impact of women's partners was not to equalise out these class inequalities but to reinforce them.

5.3

To begin with, women's male partners were likely to be in the same occupational class as the women. The majority of part-time manual women's partners were manual workers too whilst the partners of professional women were also professional workers (Table 5a). That some couples were dual working class and others were dual middle class impacted clearly on their earnings, with part-time manual women living in the lowest income households. Therefore, it was not that these women lived in male-breadwinner households where the men's wages were so high that they could just take a poorly paid job for 'pin money'. The women were contributing only a small proportion to the couples' total waged incomes. Yet because their partners were extremely low waged, the women's jobs were still no doubt a key source of household income. The extent of the low wages of these couples is illustrated by the fact that their combined monthly earnings were less than those of the partners of part-time professional women. If any women could be seen to have reduced their hours because the couples could afford it, then it would surely be these professional part-timers and not the women in the dual working class couples^[11]. Nevertheless professional part-timers were hardly earning 'pin money' and, in fact, their earnings from working part-time in higher status jobs were greater than those of women working full-time hours but in manual occupations.

Table 5: Male partner's occupational status and couples' earnings. Dual employee households where woman is aged 18 to 59

			Percentages
	Part-time			Full-time			All

	Prof.	Clerical	Manual	Prof.	Clerical	Manual	Part-time	Full-time	Total
a) Male partner's occupations:
Managerial/profes/ass. prof.	65	41	29	63	39	30	40	46	43
	(15)	(11)	(16)	(30)	(18)	(9)	(43)	(57)	(100)
Clerical	4	7	7	4	8	9	6	6	7
	(6)	(14)	(26)	(12)	(24)	(17)	(46)	(54)	(100)
Manual	31	52	64	33	54	61	54	48	50
	(6)	(12)	(31)	(14)	(21)	(16)	(49)	(51)	(100)


b) Couples' earnings:
Median monthly gross earnings:					Ł
Dual employee couple total	2555	1987	1484	2980	2238	1834	1790	2347	2097
Woman employee	754	418	277	1451	924	720	373	1001	721
Male employee partner	1801	1569	1207	1529	1314	1114	1417	1346	1376

Women's earnings as % couple total	30	21	19	49	41	39	21	43	34

N	126 (10)	149 (12)	298 (24)	258 (21)	246 (20)	163 (13)	573 (46)	667 (54)	1240 (100)

Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

5.4

Examining diversity amongst part-timers by occupational class has demonstrated that not only were part-time manual women lower waged themselves, they also lived with low waged partners in severely low-waged dual working class couple households. In addition, a class based analysis reveals that they differed from other part-timers in their class upbringing. This is addressed in the next section which asks; are part-time manual jobs simply working class jobs for working class women with children?

5.5

Part-time manual work - working class jobs for working class mothers?
To explain why some part-timers entered the lowest status jobs, it is important to look at women's educational levels as they are strong determinants of occupational attainment. Women's education was analysed via the highest level achieved, which was measured in terms of academic and vocational qualifications. Individuals were then grouped according to whether they had high (university) level, intermediate or low levels of education (see Appendix 3). Examining the qualifications of the six groups of women, it is clear that domestic responsibilities may have led female employees into part-time jobs, but women with fewest qualifications entered into the lowest quality part-time jobs available.

5.6

Overall, full-timers had higher attainment than part-timers^[12] but differences in education by class were largest amongst the part-timers. Part-timers in manual jobs had by far the lowest levels of education of the six employee groups and, in particular, they had substantially fewer qualifications than their professional counterparts (Table 6a). A possible interpretation could be that part-time manual women were qualitatively different to these other women, had never been committed to work, had not pursued qualifications because of this, and so were concentrated in low status part-time jobs. However, the locus of control which is inherent in this interpretation is questionable. Rather than suggesting that these women had voluntarily prioritised investments other than formal education as they just wanted part-time jobs (and any part-time jobs at that), women's background class was examined. A working class upbringing is known to be associated both with lack of educational success and with entry into working class jobs.

5.7

Operationalising social class is a notoriously problematic endeavour (for example, see the debates over the meaning and usefulness of 'class' between O'Reilly and Rose, Prandy, and Blackburn1998, in Work, Employment and Society, 12, 4, ), but women's class is a particular concern. Much of the discussion over measuring women's class that has taken place within British sociology has been in reaction to the approach advocated by John Goldthorpe (1983). Located within the Weberian tradition, this approach defined class by market position and, furthermore, took the household as the unit of class analysis. For Goldthorpe then, a family would be placed in a class schema based on the class position of the male head of the household. This was because most married women have very interrupted labour market careers, often dominated by part-time working, and so in this approach a wife's market position does not determine the class position of the household. Similarly, father's class indicates class background. This focus on the male head of the household has been roundly criticised and a number of alternative household based approaches were devised (see for example the 'joint' approach of Britten and Heath 1983). However, Stanworth (1984) argued that it is vital to move on to examine individuals within households to capture women's own class positions, and this individual approach is adopted here (for debates over women and class see Abbot and Sapsford 1987; Crompton 1998; Marsh 1986; Prandy and Blackburn 1997; Reay 1998; Roberts 1993). To indicate women's class upbringing, the occupational details of their parents were required and a BHPS question was used which asked respondents 'when you were aged fourteen, what occupations did your parents have?'^[13]. Parents' occupations were grouped into various socio-economic categories (see Table 6b and 6c).

5.8

Examining fathers first, over half the manual part-timers had been brought up in families where their father was a manual worker. Conversely, substantially more of the professional than the manual part-timers had professional fathers when they were aged fourteen (Table 6b). The results on mothers' occupational class were less distinct, partly because very large numbers had not been in paid work when their daughters were fourteen. However, more mothers of manual than professional part-timers had been outside the labour force, and the working mothers of part-time manual women were slightly more likely to have been in manual jobs (Table 6c). Overall, the largest proportion (one third) of the daughters of manual parents were manual part-timers, and one third of professional parents' daughters were professional full-timers. Age was controlled for as it was expected that mothers of younger cohorts would have higher employment rates and that both parents of the younger cohorts would be less likely to work in manual jobs, reflecting the general feminisation shifts in the British labour market and the growth of non-manual jobs (Brown 1997). Yet for all cohort groups of part-timers, manual women were the group most likely to have had manual parents.

Table 6: Levels of education and class background of female employees aged 18 to 59

			Percentages
	Part-time			Full-time			All

	Prof.	Clerical	Manual	Prof.	Clerical	Manual	Part-time	Full-time	Total

a) Women's educational level
High	83	32	24	80	38	40	36	55	47
	(14)	(7)	(12)	(38)	(15)	(14)	(33)	(67)	(100)
Intermediate	12	57	42	16	54	35	40	34	37
	(3)	(16)	(27)	(10)	(28)	(16)	(45)	(55)	(100)
Low	5	11	34	4	7	25	23	41	16
	(2)	(7)	(51)	(5)	(8)	(26)	(60)	(40)	(100)

b) Fathers' occupation¹:
Managerial/professional	39	19	13	36	19	15	19	24	22
	(16)	(10)	(15)	(34)	(15)	(10)	(41)	(59)	(100)
Intermediate non-manual	7	7	4	8	8	3	5	7	6
	(10)	(14)	(17)	(27)	(24)	(8)	(41)	(59)	(100)
Junior non-manual	7	6	5	6	8	4	5	6	6
	(10)	(11)	(23)	(21)	(25)	(10)	(44)	(56)	(100)
Manual	36	47	56	31	49	52	50	42	46
	(7)	(11)	(32)	(14)	(19)	(16)	(38)	(62)	(100)
Other/not in paid work	12	21	23	19	16	26	22	20	20
	(5)	(12)	(30)	(20)	(15)	(18)	(47)	(53)	(100)

c) Mothers' occupation¹:
Managerial/professional	5	4	3	6	4	4	3	5	4
	(11)	(11)	(17)	(32)	(17)	(13)	(38)	(62)	(100)
Intermediate non-manual	17	6	6	14	11	9	8	12	10
	(16)	(7)	(15)	(31)	(19)	(13)	(38)	(62)	(100)
Junior non-manual	23	23	20	23	27	21	21	24	23
	(9)	(12)	(24)	(21)	(21)	(13)	(44)	(56)	(100)
Manual	17	18	20	10	16	21	19	15	17
	(9)	(12)	(32)	(13)	(17)	(18)	(53)	(47)	(100)
Other/not in paid work	39	49	52	47	43	47	49	45	47
	(7)	(9)	(29)	(23)	(17)	(15)	(45)	(55)	(100)

N	168 (8)	209 (10)	499 (24)	467 (22)	399 (19)	355 (17)	876 (42)	1221 (58)	2097 (100)

* Socio-economic group of women's parents when the women were aged 14.
Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

5.9

These results on women's class upbringing do not support the idea that part-time manual workers prioritised domestic investments and entered the jobs which their lack of education allowed. Instead they support the proposal that women's low status jobs were related to their largely working class background and reinforce the suggestion that part-time manual jobs are working class jobs for working class women who have children. Part of this would be that a lack of qualifications served to force women into such poor quality jobs when they needed to reduce their hours for child-care reasons. However, even when qualifications are controlled for, the daughters of manual parents were substantially more likely to be in manual employment than the other women^[14]. Therefore, this cross-sectional study has shown that gender reasons can help us to understand why women in the British labour market may need to reduce their hours of employment, but knowledge of women's class background casts light on the quality of the part-time jobs which these women can then take.

5.10

Further research into the working trajectories of women from different classes would help illuminate the processes by which they entered into these jobs. A key question is how working and middle class mothers become over-concentrated at opposite poles of the part-time labour market. With a longitudinal analysis it would be possible to assess whether, for example, mothers from working class backgrounds had been working full-time in low status occupations before their need for reduced hours. Was their move into part-time employment into a similar status job or was it downwardly mobile? The latter seems more likely as, in general, manual occupations offer little opportunity for women to reduce their hours without changing jobs. Instead women have to enter jobs which are already created as part-time, and most of these are in very low paying 'sales' and 'other' work at the bottom pole of the part-time labour market.

5.11

A longitudinal analysis would help to explore how women enter the other part-time pole too, and to see whether part-time professional women had been full-time professional workers before. As women from middle class homes enter higher status and higher paying jobs, it would be expected that, on the one hand, they would have less need to reduce their hours if they had child-care responsibilities. Such women are more able to take well-remunerated maternity leave or to be able to afford private child-care for example. On the other hand, there are more possibilities for higher status women to reduce their hours in their current job and so they can avoid downward mobility into the lower status part-time labour market. Retaining their full-time equivalent working conditions, these women would form the top pole of the part-time labour market and so have more potential to increase their hours when their child-care responsibilities became less demanding. Therefore a longitudinal analysis of career trajectories is the next step in understanding the ways in which class shapes the quality of the part-time and full-time jobs taken by women from working class and middle class backgrounds.

Conclusions

6.1

The major research question stimulating this paper has been why do women work part-time in inferior part-time jobs. There has been sociological debate concerning, for example, whether female part-timers are women who choose to work part-time voluntarily to let them prioritise the home or whether domestic constraints have meant so many women with children work part-time in Britain. Issues of gender and gender roles have been fundamental to both sides of this discussion. The general conclusions have been that despite their particular gender roles, part-timers are not so markedly different from full-timers in their work attitudes and preferences, and there is also heterogeneity amongst part-timers. Yet although much of the research demonstrating these points has studied women working in high status occupations, there has been little interest in discussing class diversity amongst women employees. Therefore, this paper has analysed heterogeneity in the female part-time labour market linked to women's current occupational class and their class upbringing.

6.2

The paper used BHPS data to explore diversity in the quality of women's part-time jobs, and it was clear that the manual workers faced the most distinct and severe disadvantage in the part-time labour market. They were employed in the lowest status employment with the shortest hours, the lowest hourly wage rates and least company pension coverage. However, in questioning why women worked here, little explanation could be found in their gender roles. Although female part-timers in general differed from full-timers in their 'female caring' roles, manual workers were not very distinct from other part-timers on these indicators. The next part of this question asked whether part-time manual women could afford to take such poorly paid part-time jobs because they were supported by a high earning male-breadwinner. Unfortunately for part-time manual women, this was far from true because their partners were disproportionately likely to be in low waged manual jobs too. The polarisation which was apparent in part-timers' own experiences was reinforced when their partners were examined, and the class inequality which existed among the women was not reduced but reinforced in a household context.

6.3

A major part of the reason that part-time manual women were in low status jobs was no doubt because they had very low levels of education. However, the notion that they had voluntarily opted out of education and out of decent jobs was rejected. Instead, their low educational attainment and low status jobs were traced back to their coming disproportionately from working class backgrounds. Rather than choosing this education and work, it is more plausible that these women's employment careers have been constrained by the particular educational and other opportunities which were open to them as working class girls, and which restricted their job choices in later life.

6.4

To summarise then, at one pole of the part-time labour market stood professional mothers who were well paid, well educated women who had been brought up in middle class homes and who lived as adults with highly paid, professional partners. At the other pole were mothers who had been brought up in working class households, had low levels of education, were poorly paid and had partners in low paid manual jobs too. As a result of this diversity, we cannot lump together into one unified group, women working part-time in manual jobs and part-timers in managerial/professional and other non-manual occupations, and then talk sensibly about 'female part-time employment'. There are a wide range of job characteristics and a wide range of people hidden behind the part-time label. Female part-timers in Britain are most similar in that having dependent children has led to their working part-time. Therefore, and rightly, gender-based accounts of women's part-time employment have resulted. But class divides the female labour force and the experience of gender impacts more heavily on certain groups of women, specifically on those with least resources to moderate its effects. Bradley (1996) was quoted earlier as stating that the weight of gendered inequality is disproportionately felt at the bottom of the class hierarchy. Yet, given the greater degree of class diversity amongst the part-timers, it would seem that class disadvantage is especially apparent among women with caring obligations and that the weight of class inequality is disproportionately felt when women are in the most intense gendered caring roles.

6.5

In conclusion, despite the similarity in the impact of gender roles on women's working hours, gender cannot account for the wide diversity in female part-timers' (and full-timers') employment. It is necessary to incorporate other dimensions of stratification and, by doing so, it becomes clear that women go into the lowest status jobs only in specific social contexts. As a result, it is meaningless to look at women's employment choices without reference to the opportunities and constraints women from different classes have encountered and continue to experience in their educational and working lives.

Notes

¹See Mincer and Polacheck (1974), Becker (1985). For a discussion of the main economic approaches to part-time employment see Tam (1997). Myrdal and Klein (1956) were the first to discuss women's two roles.

²For a debate over how 'meritocratic' is Britain see for example, Saunders (1995), Lampard (1996) and Marshall and Swift (1996).

³Even though the classic study by Willis (1977) of working class school-boys shows that a number of boys did in fact 'opt out' of school saying that they did not need an education for their future working class lives, the resulting conclusions centred around issues of exclusion and not preference. Notably, there has been no subsequent argument that working class adult men who are in low paid jobs or who are unemployed are, as a result, grateful slaves who brought it all on themselves.

⁴Other research has detailed the constraints which mean that women of certain minority ethnic groups work full-time rather than part-time (Dale and Holdsworth 1998; Holdsworth and Dale 1997; Reynolds 1998). Unfortunately, due to the sample size, it was not possible to split the six groups of women into different ethnic or 'racial' groups.

⁵Proxy respondents, where another person provided their details, are excluded.

⁶See Hakim (1998b) on the problems with this definition of part-time work, and also Warren and Walters (1998).

⁷That non-manual full-timers were more likely to be 'managers and administrators' than part-timers supports Crompton and Harris's (1998a) argument that managerial occupations have restricted the introduction of part-time employment for women much more so than professional occupations.

⁸The 1999 Budget allowed women earning below the NIC but at least Ł30 a week to be entitled to Maternity Allowance (DSS 1999a) and future changes to NICs are outlined in DSS (1999b).

⁹This also supports the idea that even though they do not face disadvantage in working part-time, many women from minority ethnic groups face a different hours disadvantage in the labour market than white women. Although there were only a small number of women from minority ethnic groups in the sample of employees, these women were over-concentrated in full-time manual jobs. Other research using Census data (Dale and Holdsworth 1998) found that Indian, Bangladeshi and Pakistani female employees were over-concentrated in semi-skilled factory work whilst qualitative studies (Reynolds 1998) have shown that working class African-Caribbean women work very long full-time hours.

¹⁰Wage disadvantage can be the result of differences in, for example, the age or level of education between the part-time and the full-time samples. A standard OLS wage regression confirmed that, even after controlling for such supply factors, for female employees there was still a disadvantage associated with working part-time rather than full-time.

¹¹This is not to suggest that the incomes of both partners are distributed equally throughout the household, and that women have equal access to their male partner's earnings (see Glendenning and Millar 1991).

¹²A regression model was constructed with part-time/full-time as the dependent variable for the sample of female employees. This variable was regressed against a number of independent variables such as age, marital status, motherhood and so forth (in addition, a number of labour market indicators of high part-time employment were also incorporated such as whether the firm was a small private sector firm, whether the occupation was woman-dominated and so on to combine both supply and demand effects in one model). The results indicated that a relationship still existed between working part-time and education, after controlling for these other factors.

¹³Whilst data from Wave E of the BHPS are the focus of this paper, this question about parents' occupations was asked of the respondents in Wave A.

¹⁴For example, of the women with low education, 82% of the daughters of manual fathers were also manual workers compared with 47% of women whose fathers were in managerial/professional posts. Of the women with high levels of education, 65% of the daughters of professional fathers were in professional/managerial jobs, compared with 45% of manual workers' daughters.

Appendix 1: Usual wages in the BHPS

In the BHPS, a derived variable converts usual wage or salary in current main job, pre-tax and deductions, to a monthly amount. This variable has been constructed as follows: if last gross pay was 'the usual', this is used. If last gross pay is missing, but net pay was present and this was given as 'usual', gross pay was estimated from net pay using information about marital status, partner's activity and pension scheme membership. If the last net payment mentioned was not 'usual' pay but 'usual payment' was given as gross then this was used. Finally, if 'usual payment' was net, this was converted from the last net monthly payment (Taylor et al. 1991: 1-336).

Appendix 2: Supply characteristics of female employees aged 18 to 59

			(Percentages)
	Part-time			Full-time			All

Column %	Prof.	Clerical	Manual	Prof.	Clerical	Manual	Part-time	Full-time	Total
a) Age:
18-24	5	5	7	10	24	26	6	19	14
25-34	27	26	25	37	32	26	26	32	29
35-44	34	37	29	28	18	23	32	23	27
45-54	28	26	29	22	23	20	28	22	25
55-59	6	5	9	3	3	5	8	4	5

b) Marital status:
Married	79	82	74	51	50	43	77	49	61
Living as couple	8	5	7	14	20	21	7	17	14
Widowed/separated/divorced	5	8	12	13	9	11	9	11	10
Single	8	5	7	22	21	25	7	23	16

c) Number of dependent¹ children
0	39	37	43	74	81	76	45	77	62
1	22	27	25	14	12	13	25	13	18
2	30	29	23	9	6	10	26	8	16
3+	9	7	9	3	1	1	9	2	5

d) Age of youngest dependent child²
0-4	43	37	30	28	27	26	34	27	32
5-11	42	43	43	38	33	31	43	34	40
12-18	15	20	27	34	40	44	23	39	29

N	168 (8)	209 (10)	499 (24)	467 (22)	399 (19)	355 (17)	876 (42)	1221 (58)	2097 (100)

¹ Dependent = children aged under 16 or aged 16-18 and in school or non-advanced further education, not married and living with the parent.
² Of the women who have dependent children.
Note: Row percentages are in parentheses. Due to rounding, percentages may not add up to 100.
Source: British Household Panel Survey (BHPS), Wave E, 1995.

Appendix 3: Level of educational attainment: highest educational qualification

High
Higher degree
First degree
Teaching
Other higher
Nursing

Intermediate
A level
O level

Low
Commercial, no O level
CSE grade 2-5
Apprenticeship
Other
None

This follows the International Standard Classification of Education (ISCED), where secondary and above level education is divided into second level (stage one and stage two) and third level or university/equivalent education (OECD 1996).

Acknowledgements

: The paper was developed from research funded by ESRC grant number R00429334371. The BHPS data were made available through the ESRC Data Archive. The data were originally collected by the ESRC Research Centre on Micro-social Change at the University of Essex. Neither the original collectors of the data nor the Archive bear any responsibility for the analyses or interpretations presented here. Many thanks are due to Fiona Devine, Ken Roberts, Sally Walters and Derek Ward for very helpful advice on the research of the paper, and the comments of Liz Stanley and the two anonymous reviewers of Sociological Research Online were gratefully received.

References

ABBOTT, P.A. and SAPSFORD. R.J. (1987) Women and Social Class, London: Tavistock. BECKER, G. (1985) 'Human capital, effort and the sexual division of labour', Journal of Labour Economics, Vol. 3, pp. S33-S58.

BHOPAL, K. (1998) 'How gender and ethnicity intersect: the significance of education, employment and marital status', Sociological Research Online, Vol. 3, No. 3, .

BRADLEY, H. (1996) Fractured Identities. Changing Patterns of Inequality, Cambridge: Polity.

BRITTEN, N and HEATH, A. (1983) 'Women, men and class analysis' in E. Garmarnikow et. al. (editors.) Gender, Class and Work, London: Heinemann.

BROWN, R. (1997) 'Introduction: work and employment in the 1990's' in R.K. Brown (editor) The Changing Shape of Work, London: MacMillan Press.

BURCHELL, B.J., DALE, A. and JOSHI, H. (1997) 'Part-time work among British women' in H.P. Blossfeld and C. Hakim (editors) Between Equalisation and Marginalisation: Women Working Part-time in Europe and the United States of America, Oxford: Oxford University Press.

CAVEN, V. (1998) 'Building on the middle ground: alternatives to polarisation in women's professional employment', Paper presented at Gender, Work and Organisation Conference, January.

COOK, J. (1998) 'Flexible employment: implications for gender and citizenship in the European Union', New Political Economy, Vol. 3, No. 2, pp. 261-277.

CROMPTON, R. (1997) Women and Work in Modern Britain, Oxford: Oxford University Press

. CROMPTON, R. (1998) Class and Stratification. An Introduction to Current Debates, Cambridge: Polity Press.

CROMPTON, R. and HARRIS, F. (1998a) 'Explaining women's employment patterns: 'orientations to work' revisited', British Journal of Sociology, Vol. 49, No. 1, pp. 118-136.

CROMPTON, R. and Harris, F. (1998b) 'A reply to Hakim', British Journal of Sociology, Vol. 49, No. 1, pp. 144-147.

CROMPTON, R. and LE FEUVRE, N. (1996) 'Paid employment and the changing system of gender relations: a cross-national comparison', Sociology, Vol. 30, No. 3, pp. 427-445. CROMPTON, R and SANDERSON, K. (1990) Gendered Jobs and Social Change, London: Unwin Hyman.

DALE, A and HOLDSWORTH, C. (1998) 'Why don't ethnic minority women in Britain work part-time'? in J. O'Reilly and C. Fagan (editors) Part-time Prospects: an International Comparison of Part-time Work in Europe, North America and the Pacific Rim, London: Routledge.

DEVINE, F. (1994) 'Segregation and supply: preferences and plans amongst 'self-made' women', Gender, Work and Organisation, Vol. 1, No. 2, pp. 94-109.

DEX, S. (1987) Women's Occupational Mobility, London: MacMillan Press Ltd.

DEX, S. (1988) Women's Attitudes towards Work, Basingstoke: MacMillan.

D.S.S. (1999a) 'Miscellaneous Changes', Welfare Reform and Pensions Bill - Factsheet 8, .

D.S.S. (1999b) 'National Insurance Contributions (NICs)', Welfare Reform and Pensions Bill - Factsheet 7, .

EGERTON, M. (1997) 'Occupational inheritance: the role of education and gender', Work, Employment and Society, Vol. 11, No. 2, pp. 263-283.

FAGAN, C. (1999) 'The Temporal Re-organisation of Employment and the Rhythm of Household Schedules' Paper presented at ESA Conference Will Europe Work, University of Vrije, Amsterdam, August.

FAGAN, C. and RUBERY, J. (1996) 'The salience of the part-time divide in the European Union', European Sociological Review, Vol. 12, No. 3, pp. 227-250.

FAGAN, C. and O'REILLY, J. (1998) 'Conceptualising part-time work: the value of an integrated comparative perspective' in J. O'Reilly and C. Fagan (editors) Part-time Prospects: an International Comparison of Part-time Work in Europe, North America and the Pacific Rim, London: Routledge.

GINN, J. and ARBER, S. (1998) 'How does part-time work lead to low pension income?' in J. O'Reilly and C. Fagan (editors) Part-time Prospects: an International Comparison of Part-time Work in Europe, North America and the Pacific Rim, London: Routledge.

GINN, J. and ARBER, S. (1996) 'Patterns of employment, gender and pensions: the effect of work history on older women's non-state pensions', Work, Employment and Society, Vol. 10, No. 3, pp. 469-490.

GINN, J., ARBER, S., BRANNEN, J., DALE, A., DEX, S., ELIAS, P., MOSS, P., PAHL, J., ROBERTS, C. and RUBERY, J. (1996) 'Feminist fallacies: a reply to Hakim on women's employment', British Journal of Sociology, Vol. 47, No. 1, pp. 167-174.

GLENDENNING, C and MILLER, J. (editors)(1991) Women and Poverty in Britain, Brighton: Wheatsheaf Books Ltd.

GLOVER, J. and ARBER, S. (1995) 'Polarisation in mothers' employment', Gender, Work and Organisation, Vol. 2, No. 4, pp. 165-179.

GOLDTHORPE, J.H. (1883) 'Women and class analysis', Sociology, Vol. 17, pp. 465-488.

HAKIM, C. (1991) 'Grateful slaves and self-made women: fact and fantasy in women's work orientations', European Sociological Review, Vol. 7, No. 2. pp. 101-121

HAKIM, C. (1995) 'Five feminist myths about women's employment', British Journal of Sociology, Vol. 46, No. 3, pp. 428-455.

HAKIM, C. (1996a) Key Issues in Women's Work: Female Heterogeneity and the Polarisation of Women's Employment, London: The Athlone Press Ltd.

HAKIM, C. (1996b) 'The sexual division of labour and women's heterogeneity', British Journal of Sociology, Vol. 47, No. 1, pp. 178-188.

HAKIM, C. (1997) 'A sociological perspective on part-time work' in H.P. Blossfeld and C. Hakim (editors) Between Equalisation and Marginalisation: Women Working Part-time in Europe and the United States of America, Oxford: Oxford University Press.

HAKIM, C. (1998a) 'Developing a sociology for the twenty-first century: preference theory', British Journal of Sociology, Vol. 49, No. 1, pp. 137-143.

HAKIM, C. (1998b) Social Change and Innovation in the Labour Market, Oxford: Oxford University Press.

HALSEY, A., HEATH, A. and RIDGE, J.M. (1981) Origins and Destinations, Family, Class and Education in Modern Britain, Oxford: Oxford University Press.

HEERY, E. (1998) 'Campaigning for part-time workers', Work, Employment and Society, Vol. 12, No. 2, pp. 351-366.

HOLDSWORTH, C. and DALE, A. (1997) 'Ethnic differences in women's employment', Work, Employment and Society, Vol. 11, No. 3, pp. 435-457.

HORRELL, S., RUBERY, J. and BURCHELL, B. (1990) 'Gender and skills', Work, Employment and Society, Vol. 4, No. 2, pp. 189-216.

LAMPARD, R. (1996) 'Might Britain be a meritocracy?: a comment on Saunders', Sociology, Vol. 30, pp. 387-393.

LEWIS, J. (1992) 'Gender and the development of welfare regimes', Journal of European Social Policy, Vol. 2, No. 3, pp. 159-173. MACEWAN SCOTT, A. and BURCHELL, B. (1994) 'And never the twain shall meet?' in A. MacEwan Scott (editor) Gender Segregation and Social Change, Oxford: Oxford University Press.

MARSH, C. (1986) 'Social class and occupation' in R.G. Burgess (editor) Key Variables in Social Investigation, London: Routledge.

MARSHALL, G. and SWIFT, A. (1996) 'Merit and mobility: a reply to Peter Saunders', Sociology, Vol. 30, pp. 375-386.

MARSHALL, G., ROSE, D., NEWBY, H and VOGLER, C. (1988) Social Class in Modern Britain, London: Unwin Hyman.

MARTIN, J. and ROBERTS, C. (1984) Women and Employment: A Lifetime Perspective, London: HMSO.

MCRAE, S. (1995) Part-time Work in the European Union: The Gender Dimension, Luxembourg: Office for the Official Publications of the European Communities. MINCER, J. and POLACHECK, S. (1974) 'Family investment in human capital: earnings of women', Journal of Political Economy, Vol. 82, No. 2.

MYRDAL, A. and KLEIN, V. (1956) Women's Two Roles. Home and Work, London: Routledge and Kegan Paul Ltd.

OECD. (1996) Education at a Glance - OECD Indicators, Paris: OECD.

O'REILLY, K. and ROSE, D. (1998) 'Changing employment relations: plus ca change, plus c'est la meme chose? Reflections arising from the ESRC review of Government social classifications', Work, Employment and Society, Vol. 12, No. 4, pp. 713-734.

PRANDY, K. and BLACKBURN, R. (1997) 'Putting men and women into classes: but is that where they belong?', Sociology, Vol. 31, pp. 143-52.

PROCTOR, I and PADFIELD, M. (1999) 'Work orientations and women's work: a critique of Hakim's theory of the heterogeneity of women', Gender, Work and Organisation, Vol. 6, No 3, pp. 152-162.

REAY, D. (1998) Rethinking social class: qualitative perspectives on class and gender', Sociology, Vol. 32, No. 2, pp. 259-276.

REYNOLDS, T. (1998) 'African-Caribbean mothering, paid work and identity', Paper presented at ESRC Seminar Series Parenting, Motherhood and Paid Work, 10-11 September, University of Bradford.

ROBERTS, H. (1993) 'Women and the class debate' in D. Morgan and L. Stanley (editors) Debates in Sociology, Manchester: Manchester University Press.

ROWLINGSON, K., WHYLEY, C., and WARREN, T. (1999) Wealth in Britain. A Life-cycle Perspective, London: PSI.

RUBERY, J. (1998) 'Part-time work: a threat to labour standards?' in J. O'Reilly and C. Fagan (editors) Part-time Prospects: an International Comparison of Part-time Work in Europe, North America and the Pacific Rim, London: Routledge.

RUBERY, J. and TARLING, R. (1988) 'Women's employment in declining Britain' in J. Rubery (editor) Women and Recession, London: Routledge and Kegan Paul.

RUBERY, J., HORRELL, S. and BURCHELL, B. (1994) 'Part-time work and gender inequality in the labour market' in A. MacEwan Scott (editor) Gender Segregation and Social Change, Oxford: Oxford University Press.

RUBERY, J., SMITH, M., FAGAN, C. and GRIMSHAW, D. (1998) Women and European Employment, London: Routledge.

SAUNDERS, P. (1995) 'Might Britain be a meritocracy?', Sociology, Vol. 29, pp. 23-41.

SCHEIBL, F. (1998) 'Self made women: agency, time and constraint', Paper presented at Gender, Work and Organisation Conference, January.

STANWORTH, M. (1984) 'Women and class analysis', Sociology, Vol. 18, pp. 159-170.

SWIFT, A. and MARSHALL, G. (1997) 'Meritocratic equality of opportunity: economic efficiency, social justice or both?', Policy Studies, Vol. 18, No. 1, pp. 35-48.

TAM, M. (1997) Part-time Employment: a Bridge or a Trap?, Aldershot: Avebury.

TAYLOR, M. (1998) 'The modernisation of Britain's tax and benefit system. Number Two. Work incentives', .

TAYLOR, M.F., BRICE, J., BUCK, N. and PRENTICE, E. (1991) BHPS User Manual Volume 8: Introduction, Technical Report and Appendices. Colchester: University of Essex.

WALSH, J. (1999) 'Myths and counter-myths: an analysis of part-time female employees and their orientations to work and working hours', Work, Employment and Society, Vol. 13, No. 2, pp. 179-203.

WARREN, T. and WALTERS, P. (1998) 'Appraising a dichotomy: a review of the use of 'part-time/full-time' in the study of women's employment in Britain', Gender, Work and Organisation, Vol. 5, No. 2, pp. 102-1181.

WILLIS, P. (1977) Learning to Labour. How Working Class Kids Get Working Class Jobs, London: Saxon House.

WITZ, A. (1992) Professions and Patriarchy, London: Routledge.