Facets of Gender

INTRODUCTION Facets of gender colour our life from the day we are born until the day we die. When the proud parents announce the birth of a child, the sex of the child is what is first conveyed to relatives and friends who happily bring gifts in the form of baby clothes of the ’right’ colour. As we grow up, we learn to separate between girls’ and boys’ toys and games. Not surprisingly, children also learn rather early in life that there are differences between male and female behaviour (Fagrell 2000; Wernersson 1977). The first experiences of gender differences in adult behaviour generally occur in the family of origin and in the household. Although Swedish women and men on the whole work the same total number of hours, women do more unpaid work than men do (e.g., Flood & Gråsjö 1997; Nermo 1994; Statistics Sweden 2003). Women spend more hours doing housework than men do, and the difference is more pronounced in families with small children (Ahrne & Roman 1997; Flood & Gråsjö 1997; Hörnqvist 1997; Statistics Sweden 2003). The fact that women often take the primary responsibility for childcare and housework affects their opportunities in the labour market. Employers, aware of women’s larger household and care burden, tend to see women as more ‘risky’ labour than men, and as a consequence, they may choose not to invest in women to the same extent as in men (Becker 1975; Lazear & Rosen 1990). Interactive processes make gender differences and inequalities in the household difficult do disentangle from differences and inequalities in the labour market (cf., Acker 1990).

In this thesis, the first two papers focus on how gender differences are expressed within the household. In Paper I, I study gender differences in the housework young Swedish boys and girls perform. The analyses show that even among these children, age 10 to 18, the division of work is gendered. The recent and rich data – The Swedish Child Level of Living Survey 2000 – which include first hand information from both parents and children are used to study the extent to which parents’ division of work is related to their children’s work. In Paper II, the focus is on theories behind the division of housework in adult couples. The purpose is to study the relative importance of the resource-

1

bargaining perspective and the ‘doing gender’ approach for predicting the division of housework among adult women and men in Sweden and the United States. The longitudinal nature of the American Panel Study of Income Dynamics (PSID) and the Swedish Level of Living Survey (Levnadsnivåundersökningen; LNU) allows comparison both across countries and over time, an unusual strength in these types of studies.

In Paper III, the public and private spheres are most clearly linked when differences in labour market outcomes after divorce are in focus. The work and family history data of the LNU 1991 are used to estimate hazard regression models of the competing risks of changing to a job of higher versus lower occupational prestige. The results show that women’s ability to improve their job situation, i.e. to change to a job of higher prestige after divorce, is dependent on their investment in paid labour during the time they were cohabiting or married. In the fourth and final paper, I also focus on advancement opportunities, but this time in terms of gender differences in formal on-the-job training and its consequences for the gender wage gap. By using the Swedish Survey of Living Conditions (Undersökningen av levnadsförhållanden, ULF) from the mid 1990s, I am able to distinguish between different types of training, such as firm-specific training, industry-specific training and general training. A fourth type of training that cuts across the other three is training that increases promotion opportunities. A distinction between different types of training is important, not least with regard to estimating gender differences in the returns on training.

The purpose of this introductory chapter is to provide a background to the papers included in the thesis. First, the Swedish context is depicted. The description focuses on gender equality and gender differences in the labour market and in the home in Sweden. A brief account of some historically significant political developments and changes is also given. After the section on the Swedish context is a discussion of theories concerning the division of work in the household, followed by a brief account of two theories of discrimination, i.e. taste discrimination and statistical discrimination. The theory part is concluded with a discussion of social closure processes and gendered organizational structures. Last but not least, the four empirical papers are summarized.

2

GENDER INEQUALITY IN THE SWEDISH CONTEXT Today, Sweden is known as a comparatively gender egalitarian country. When the United Nations in 1995 used the Gender-Related Development Index (GDI)1 to estimate the degree of gender inequality in 130 countries, Sweden was found to be the country with the fewest gender disparities (Human Development Report 1995). The fact that Sweden today, together with the other Nordic countries, is leading with regard to gender equality is due to several factors, most of which will be discussed below.

The gender equality/female employment debate in the 1960s and the introduction of separate taxation and parental leave insurance in the 1970s.

In the 1950s, female employment was exceptionally low in Sweden (see Nermo 1999, Fig. 2.1). Instead, in the 1950s and early 1960s, labour immigration was substantial. From the mid 1960s, however, the union lobby became concerned about immigration (including tourist immigration).2 Partly due to the union lobby instead stressing the importance of employing traditionally weak Swedish labour groups such as youth, women and the elderly, labour immigration ceased in the late 1960s and early 1970s (Lundh & Olsson 1994). In the early 1960s, women and men’s traditional roles were debated, and in the mid 1960s, both the political establishment and the women’s movement saw married women’s employment as crucial for gender equality (Baude 1992). Labour shortage led to a demand for political reforms that would facilitate female employment and, gradually, the incentives for women to work increased. In 1965, special (i.e., especially low) ‘women’s wages’ were formally abandoned after an agreement between the employers’ organization SAF (Svenska arbetsgivareföreningen) and the trade union federation LO (Landsorganisationen, organizing manual workers). In 1971, individual taxation was introduced, and in 1974 the parental leave insurance replaced the maternity leave system.3 Although many married women already benefited from split taxation due to a deduction for working married women that was introduced in 1952 (Elvander 1974), the reform in 1971 signalled the commitment of the government and the social democratic party to support women’s employment. This was reinforced by a strong expansion of public sector employment in service occupations, primarily care and education. Together, this led to a rapid increase in the number of gainfully

3

employed women. In 1968, about half of all married women were employed. In 1981, this share had increased to 82 percent (Axelsson 1992). The results of the expansion of the public sector for female employment were twofold. First, by putting care work in the hands of the municipalities, women could get paid for doing what they normally did without any pecuniary rewards at home. Second, as the public childcare system was gradually expanded, mothers were able to send their pre-school children to high quality state-subsidized childcare during their work hours.

The introduction of the parental leave insurance in 1974 was aimed at allowing women to combine work and childcare, as well as to increase equality between women and men in the labour market and in the household (Government Bill 1973:47). Fathers and mothers were given equal rights to take parental leave for six months with the right to return to the same employer after the leave (the latter right was also included in the maternity leave insurance). The income replacement level was set to 90 percent of the individual’s gross income (up to a ceiling). Parents who did not work at the time of the child’s birth received only a much lower flat-rate benefit. Since that time, the parental leave insurance has been extended several times, and in 2002 the mother and the father together have the right to take 480 days of leave (of which 90 days are replaced at the lower basic level). Parents can give up all but 60 days to the partner. These 60 days are an extension of what was first termed ‘the daddy’s month’ at the introduction in 1995. The replacement level has varied and since 1998 the parental leave insurance covers 80 percent of the parent’s gross income.

The parental leave insurance gives young women strong incentives to work for some time before the birth of the first child (due to the larger compensation for the employed). The right to paid leave with job security has meant that almost all mothers stay at home for a majority of the currently permitted parental-leave period, at the same time as it has accelerated the rate at which the group with earlier ‘home-maker’ characteristics returns to work after childbirth (Jonsson & Mills 2001; Rønsen & Sundström 1996). One of the aims of the parental leave insurance has thus been realized: Women’s possibilities to combine work and children have increased. The extent to which it has promoted gender inequality in the home and in the labour market is, however, questionable. Women still use the majority of all leave: in 2002 men took only

4

16 percent of all parental leave days (The National Swedish Social Insurance Board 2003). Men take a larger share of the parental leave in families where the mother has a higher education, and the higher the income the father has – up to a ceiling – the more parental leave he takes (The National Swedish Social Insurance Board 2002; Sundström & Duvander 2002). This indicates that gender equality in the home has increased the most for highly educated women and for women with a high-income husband.

Gender inequality in the labour market

A large majority of all Swedish women are gainfully employed. The economic activity rate among women 16-64 years of age was 76 percent in 2001 compared to 80 percent among men. In the same year, about 48 percent of all employed were women. Among those, 66 percent worked full-time (Statistical Yearbook of Sweden 2003). Hence as much as a third of all gainfully employed women work part-time. The fact that parents are given the right to work part-time when they have children under eight is reflected in the proportion of female part-time workers in this category: about 50 percent worked part-time in 1999 (Båvner 2001). Part-time is defined as working less than 35 hours a week and a majority of all women work long part-time, i.e. between 20 and 34 hours per week. Worth noting is that part-time workers in Sweden do not fall outside either the social security or the pension scheme. They also have the same legal rights as full-time workers concerning, e.g., employment security.

The Swedish labour market is segregated by sex. This segregation is apparent across industries as well as occupations and jobs. Women more often work in the health and medical care industry, in social services and education. Men more often work in manufacturing industries, in construction and transport (see Paper IV, Appendix 1). The sex segregation by industry is mirrored in segregation by employment sector: Whereas a majority of all men work in the private sector, half of the women work in the public sector (Statistics Sweden 2002). Although 74 percent of all those employed in the public sector are women, they only represent 55 percent of the managers in this sector. In the private sector, 81 percent of all managers are men, whereas men make up 63 percent of all those employed in that sector (SOU 2003:16). In the labour market

5

at large and in the private sector in particular, women are restricted from attaining supervisory positions (Hultin 1998).

The female labour force is more concentrated in certain occupations than the male labour force is: In 2001, the 30 largest Swedish occupations employed 58 percent of all gainfully employed women compared to 37 percent of all gainfully employed men (Statistics Sweden 2002).4 Although sex segregation still is considerable, the level of segregation has declined. In 1968, 68 percent of all employed women (or men) would have needed to change occupation in order for the labour market to be sex-integrated, whereas the corresponding figure in 1991 was 53 percent (Nermo 1996).5 The sex-segregated Swedish labour market is, however, not exceptional in an international perspective (Charles 1992; Nermo 2000) (See, however, Padavic & Reskin 2002 for a somewhat divergent view).6

The sex-segregated labour market is partly a result of women and men choosing different educational tracks and as a consequence different occupations. Although women have dominated academic secondary and tertiary education in Sweden since the early 1980s – except at the post-graduate level – the horizontal sex segregation in type of educational track chosen has not decreased (Jonsson 1999). The reason for the gender difference in educational choice is, according to Jonsson (1999), partly due to the comparative advantages that girls and boys (expect to) have in different educational programmes.7 It is worth noting, however, that the expected success in a given educational area – and in a subsequent occupation – is most likely also affected by factors outside the individual’s control, such as expected difficulties in belonging to the minority sex in a given occupation (cf., Kanter 1977).8 Even when boys and girls choose the same education at the gymnasium level, they tend to occupy different jobs after education is completed (Hoem 1995; Statistics Sweden 1991). Also in predominantly female occupations, men tend to occupy the more qualified positions (cf., Hultin 2003; Statistics Sweden 1991).

Educational and occupational segregation are reflected in wage differences between women and men. In 2000, women earned 82 percent of what men earned (Statistics Sweden 2002). A large fraction of the gender difference in pay is due to women working in industries and occupations that pay less than do those that men work in (le Grand 1997; le Grand et al. 2001). Also, more women than men work part-time (see above). Controlling for age,

6

education, work hours, employment sector and occupational group, the remaining gender wage difference is 8 percent (Statistics Sweden 2002). In an international context, the Swedish gender wage gap is low (Blau & Ferber 1992; Blau & Kahn 1996; Sørensen 2001). The greatest change in the gender wage gap came about in the 1960s and 1970s when union power peaked and blue-collar trade unions promoted a ‘solidaristic wage policy’ (i.e., focussed on raising the wages for the lowest paid) in wage negotiations. Still, the decisive factor behind the reduction in the gender wage gap appeared to be market forces and the excess demand for female labour (Svensson 2003). Since the 1980s, the reduction in the gender wage gap has stagnated (le Grand et al. 2001; Svensson 2003).

Economic dependency, divorce and the distribution of work in the household.

The high female employment rate together with the comparatively narrow gender wage gap in Sweden contributes to greater gender equality in the home. Swedish women’s economic dependency on their spouse is low in an international perspective (Hobson 1990; Sørensen 2001; see also Paper II, Appendix 1). In 1995, Swedish women living with a partner and at least one child under seven years of age had an average dependence of 0.23, meaning that they relied on their spouse for 23 percent of their share of the couple’s combined earnings (Sørensen 2001). The economic dependency indicator (originally derived from Sørensen & McLanahan 1987) does not, however, take into account income from earnings-determined transfers such as employment insurance, disability insurance and parental leave schemes (Sørensen 2001). As a consequence, it overestimates Swedish women’s economic dependency. When instead the gender poverty ratio9 is estimated – and social transfer incomes are included – Swedish women’s poverty rate is actually lower than men’s (Caspar et al. 1994; Christopher et al. 2002). Overall poverty rates in Sweden are low in an international perspective, and according to this definition, only 3 percent of men and 2 percent of women are living in poverty (Christopher et al. 2002). Single mothers have a poverty rate of 5 percent (Ibid.). In an international perspective, this is an exceptionally low figure.10 The Swedish tax and transfer system is of considerable importance with regard to reducing poverty among single mothers (Christopher et al. 2002; Båvner 2001). It also plays an important

7

role in reducing the overall sex-poverty ratio (Christopher et al. 2002). Worth noting, however, is that although Swedish single mothers fare financially well in an international perspective, their life is far from problem free (see Gähler 1998). Also, recent research shows that single mothers’ financial situations worsened during the 1990s (Gähler 2001; Hobson 2003; Hobson & Takahashi 1997).

All in all, however, as is obvious from the above discussion, Swedish women have a comparatively high capacity to form and maintain an autonomous household – i.e. to survive and support their children (if any) without having to marry to gain access to a breadwinner’s income (Orloff 1993). Perhaps partly as a consequence, the divorce rate in Sweden is high – in a 16-country comparison, Andersson (2003) showed that only the US has higher figures of separation (merging marriages with consensual unions) than does Sweden, though the Swedish figures are not so high internationally when counting only couples with children (Andersson 2002). In 2001, 13.5 of 1000 marriages in Sweden ended in divorce (Statistical Yearbook of Sweden 2003). This is about 25 times the divorce rate at the start of the century (Statistics Sweden 1990). The largest increase in the divorce rate has taken place since the late 1960s, i.e. since married women’s participation in gainful employment started to rise again (see above). There appears to be a direct link between women’s employment (i.e., their reduced economic dependency) and the divorce rate as marriage dissolution is higher in families in which the spouses make about the same contribution to the family income (Henz & Jonsson 2003).11

One of several reasons for divorce and separation can be conflicts about the division of work in the home (Wadsby & Svedin 1993). Even in families that are intact, conflicts about the division of housework are not uncommon. Ahrne and Roman (1997) found that 35 percent of women stated that they sometimes or often had disagreements with their spouse about the division of housework and 66 percent said that they sometimes or often wanted their spouse to do more housework. That conflicts about the division of housework are comparatively common is probably due not only to a general time-squeeze in two-earner families, but also to the fact that women still do the majority of all housework. At the beginning of the 1990s, cohabiting women spent about 19 hours per week in housework (Flood & Gråsjö 1997; Hörnqvist 1997; Nermo 1994). Information concerning the amount of time men spend on this type of work at

8

the same time varies between 5-6 hours (Hörnqvist 1997; Nermo 1994) and ten hours (Flood & Gråsjö 1997). 12 This means that women on average spend (at least) about twice the amount of time on housework as men do. This is also the case in 2000/01 (Statistics Sweden 2003). Worth noting, though, is that men spend more time than women on maintenance and repair work, although this work is less time consuming than housework (Statistics Sweden 2003, cf. also Paper I). When women and men are singles, they spend about the same amount of time on housework (Hörnqvist 1997).13 In light of this, the unequal division of work among cohabiting and married women and men calls for an explanation. In the next section, two overarching theories or approaches that set out to explain this unequal division of work are discussed.

DOING GENDER AND BARGAINING THEORY Both the ‘doing gender’ approach and theories about household bargaining contribute to our understanding of the division of work in the household, and women’s larger housework burden in particular (see Paper I and II). A third theory that has been used to explain various differences in women’s and men’s work and behaviour is sex role theory. This theory has, however, been abandoned by most feminist sociologists and I will initially give a brief account for why this has occurred.

Sex role was an established concept in conventional family studies in the 1960s and 1970s. As a consequence, many feminist scholars adopted the concept and only criticized the normative implications of its theoretical framework (Feree 1990). In the late 1970s and 1980s, however, several scholars started to question the usefulness of sex role theory as a tool for understanding gender differences and their genesis. The most radical criticism concerned the static nature of the theory, i.e. actors do unto others what was done to them in an infinite regress (Connell 1985). Hence sex role theory offers no explanation for historical changes. Criticism was also directed towards the theoretical and empirical problems that resulted from the term ‘sex role’. It was being used as a catch-all term for everything from structural disadvantage to implied personality traits, even though the term ‘role’ should only prescribe specific behaviours towards specific others (Lopata & Thorne 1978). As Feree (1990) summarized her review of the critique of the sex roles approach: “..the concept of “sex roles”

9

[is] rooted in socialization, internalized in individuals, and merely echoed in and exploited by other social institutions, [it] cannot encompass the actual variation in men’s and women’s lives – individually over the life course and structurally in the historical context of race and class” (Feree 1990:868).

Partly as a consequence of this, West and Zimmerman (1987) stressed the importance of gender as a routine, methodical, and recurring accomplishment in their article “Doing Gender”. Hence rather than seeing gender as something internal to the individual, gender is here seen as an emergent feature of social situations. Gender is created in social interaction and it is recurrently created by women and men whose competence as members of society is built on the very doing of gender (Ibid). West and Zimmerman (1987) emphasized the distinction between sex, sex category and gender. Whereas sex is the socially agreed upon biological criteria for classifying persons as males and females, the placement in a sex category is based on the socially required identificatory displays that proclaim one’s membership in one or the other category (for instance, gendered clothing, hairstyles and cosmetics). Gender, on the other hand, is the activity of managing situated conduct in light of normative conceptions of attitudes and activities appropriate to one’s sex category (West & Zimmerman 1987:127). Thus, doing gender involves making visible differences between boys and girls, men and women, that are not biological, natural or essential (Ibid.). However, once these differences are made visible through institutions, discourse, media images and daily practice, they are often perceived as natural and used to reinforce the essentialness of gender. An example of this is the division of housework. Even when both the woman and the man work fulltime, the woman generally spends more time doing housework than does the man (see Papers I and II for Sweden). As Fenstermaker Berk (1985) argued, household members appear to do gender as they do housework and childcare. Doing gender means engaging in behaviour at the risk of gender assessment (West & Zimmerman 1987). In virtually everything we do, we are assessed and held accountable on the basis of gender. A tangible example might again be with respect to housework. If the home is untidy and/or the kids are not clean, most people would hold the mother – rather than the father – responsible for this, regardless of whether they both work the same number of hours in gainful employment (cf., Lorber 1994:175).

10

The doing gender approach has, among other things, been criticized for having a functionalist tilt. By emphasizing the maintenance and reproduction of normative conceptions, the proponents of this approach tend to neglect countervailing processes of resistance, conflict and change (Thorne in Fenstermaker & West 2002). As Lorber (1994) argued, gendered roles change – today men’s share of the housework is greater than it used to be and more fathers are taking care of children, for example. Even so, gender is a social institution. It is one of the most important conceptions according to which human beings organize their lives. Boys and girls, men and women, are assigned different roles and responsibilities. Different life experiences lead to different feelings, motivations and behaviours. The gendering process that creates these differences is legitimated by religion, law, science, and society’s entire set of values (Lorber 1994). Consequently, even if the definitions of appropriate male and female conduct and behaviour are not as strict today as they were fifty years ago, we all still do gender consciously and unconsciously in social interactions in everyday life.

If the doing gender approach (simplified here) suggests that the division of housework in heterosexual families is the result of gender-creating processes, the household bargaining approach suggests that the division of housework is the outcome of implicit or explicit bargaining processes where one’s bargaining power is a function of one’s resources. Whereas several related perspectives could be summarized under the latter heading, I will concentrate on two of them: cooperative bargaining models (e.g., Lundberg & Pollak 1996; cf., also England & Kilbourne 1990; Hobson 1990) and the relative resource-bargaining perspective (e.g., Blood & Wolfe 1960). A third theory that has often been used with regard to explaining gender differences in the household is the theory of specialization (also referred to as the altruist model or the common preference model) (Becker 1981; 1985). According to this theory, the family is seen as a single utility maximizing unit with an altruist head who distributes resources justly among family members. A major problem with this theory, however, is that it fails to recognize power and male dominance in the family (e.g., England 1993; England & Budig 1998). Also, empirical evidence suggests that the family income pooling assumption that forms the basis of this theory must be rejected. If this assumption were credible, only total income – not the person distributing the income within the family – should play a part in family consumption. In

11

contrast to this, studies have found that children tend to do better in families in which the mother controls a larger fraction of the family resources (Haddad & Hoddinott 1994; Thomas 1990; 1994, referred to in Lundberg & Pollak 1996).

In cooperative bargaining models, the family is seen as consisting of two or more agents with distinct preferences with regard to (for instance) family consumption (Lundberg & Pollak 1996; Manser & Brown 1980; McElroy & Horney 1981). Cooperative bargaining models can be divided into external and internal threat point models (Lundberg & Pollak 1996). According to the external threat point model, also termed the divorce-threat bargaining model, when agreement is not reached simply, agent A’s chances of having things his/her way are dependent on the threat-point or the fallback options A would have if he or she were to leave the relationship. The better the fallback options, the better the negotiated outcome (see Lundberg & Pollak 1996 for a more detailed and economically advanced account of the theory). The divorce threat point, however, is not only determined by the income received by the husband and the income received by the wife, but also by environmental factors such as conditions in the remarriage market and income directed to divorced women and men (e.g., welfare payments to single mothers) (Ibid.).

Some sociologists have expressed the same concepts in slightly different language. Hobson (1990) elaborates on Hirschman’s framework of exit and voice (Hirschman 1970). According to Hobson, the more economically dependent the agent is, the weaker his/her voice (i.e., the ability to make claims in family decision making), the lower his/her earnings potential, the fewer his/her exit possibilities – and the fewer the exit possibilities, the weaker the voice (Hobson 1990). A similar approach was suggested by England and Kilbourne (1990). According to them, an agent’s power in negotiations concerning consumption and other decisions in the family is dependent on “…how much one contributes to a relationship, the ease with which one could leave the relationship and take the fruits of such contributions, the extent to which one is inclined toward self-interested bargaining, how much one’s contributions are valued by the partner, how this compares to the value the partner places on what could be had outside this relationship, and how one compares what is had within the current relationship to what could be had outside it.” (England & Kilbourne 1990: 170). Hence in practice, this

12

corresponds to a divorce-threat bargaining model in which the threat point is the maximal utility attainable outside the marriage (cf., Lundberg & Pollak 1996).

Whereas according to the divorce-threat bargaining model threat points are seen as external to the marriage, in the separate spheres bargaining model threat points are internal to the marriage (Lundberg & Pollak 1996). This means that instead of implicitly threatening to leave the relationship, one can (threaten to) withhold one’s own contributions from the spouse to some extent. This is similar to Blood and Wolfe’s (1960) relative resource-bargaining perspective, according to which the amount of domestic work performed by each spouse is determined by the distribution of relative resources between them. The more resources a person has at his/her disposal (e.g. financial, educational and social status resources), the more likely this person will be to negotiate away this work. The latter perspective does however not include external contextual factors (such as financial support to single mothers) among the factors predicting the outcome.

The theories discussed above contribute to our understanding of gender differences in unpaid work. Such work, however, only constitutes a part of all work performed by women and men. In paid work, gender differences are also prevalent, and although part of these differences can be referred to as differences in men’s and women’s preferences for different jobs (Marini & Fan 1997; Okamoto & England 1999) among other things, the discussion in the following will focus on gender discrimination in the labour market. That is, when two equally qualified individuals are treated differently solely on the basis of their gender (Blau & Ferber 1992, referring to Becker 1971).

THEORIES ON GENDER DISCRIMINATION IN THE LABOUR MARKET The most common example of a direct effect of discrimination is when unequal treatment results in women’s wages being lower than men’s even when they have the same productivity-related characteristics or qualifications (Altonji & Blank 1999). Two other examples are when a woman, in a corresponding situation, is denied access to employer-sponsored training (Blau & Ferber 1992) (a problem treated in Paper IV), or when she is restricted from getting access to authority positions in the workplace (Hultin 1998; Hultin & Szulkin 1999). Indirect effects of discrimination can occur if discrimination has feedback

13

effects such that it affects women’s behaviour in ways that result in them being less well-qualified than men are (Blau & Ferber 1992). Most theoretical and empirical work, however, has focussed on the direct effects of discrimination. This is also what I will focus on below.

Economic models of discrimination can be divided into two main classes: competitive and collective models of discrimination.14 Most theoretical work by economists has concentrated on competitive models of discrimination, i.e. models in which agents act individually (as compared with collective models in which one group acts collectively against another, cf. Reskin 1988, below). The competitive framework emphasizes two types of discrimination, the first of which is taste discrimination, meaning that at least some members of the majority group have prejudices or ‘tastes’ against interacting with members of the minority group (Becker 1971). The second is statistical discrimination by employers in the presence of imperfect information about the skills or behaviour of members of the minority group. In the following, I will concentrate on taste discrimination among employers. Tastes against associating with the minority group, however, can be held either by employers, co-workers or customers.15

Taste discrimination among employers will result in a gender wage gap when employers are willing to pay more to hire members of the preferred group (i.e., men) than they are to pay equally qualified members of the disfavoured group (women) (England 1992). The larger the number of prejudices held by employers and the stronger the intensity of their preferences, the larger the gender wage gap (Altonji & Blank 1999). As a consequence, employers that lack a taste against women will be more profitable than prejudiced employers, as the former will pay less for their – equally skilled – workforce than will prejudiced employers. In the long run, this should, according to Becker and other economists, lead to the abolition of taste discrimination and consequently the elimination of the gender wage gap as low-profit prejudiced employers are eliminated by high profit non-discriminatory employers. Because the gender wage gap still exists, the conclusion, according to Altonji and Blank (1999), is that a) either there is no taste discrimination or, b) it is not the primary form of discrimination in the labour market or, c) all potential employers are discriminators and/or, d) other factors interfere with the expansion of non-discriminating firms, such as search friction or collective action. Worth noting is that if the taste discrimination framework is extended to incorporate the

14

possibility that employers’ preferences for female workers are dependent on the type of job concerned, this leads to a theory of occupational segregation (Ibid).

The basic premise of statistical discrimination is that employers have limited information about the skills job applicants’ possess and their turnover propensity. This uncertainty is especially great when potential new employees are young and have little labour market history. If firms in these situations can use readily observable characteristics such as gender in order to statistically discriminate between in other respects equal applicants, they will do so if these observable characteristics are correlated with performance (Altonji & Blank 1999). As is obvious from this description, statistical discrimination is often assumed to be most influential in hiring decisions when the employer has little information about the individual applicant. With regard to already employed women’s and men’s assignment to employer-sponsored training programs or promotions, statistical discrimination should play a minor part, as the employer in these cases has been able to observe the employee and his/her skills for some time. Worth noting is that the quality of employers’ information may differ due to the gender of the employee (Altonji & Blank 1999; England 1992). This is what England (1992) termed error discrimination. An example of this is if a male manager is a worse judge of female employees than of male employees. One reason for this may be that social networks tend to run along gendered lines and personal contacts are an important source of information in the workplace (cf., Hultin & Szulkin 1999).

It has often been suggested that the occurrence of gender discrimination is indicated by the ‘unexplained gap’ in wage regressions, i.e. the difference in wages after controlling for a host of personal and job characteristics (Altonji & Blank 1999). Reliance on a statistical residual, however, does leave the approach open to criticism as to whether all relevant variables were actually included in the regression (Blau & Kahn 2000). As a consequence, the ‘true’ gender discrimination effect could either be overestimated or underestimated (see Altonji & Blank 1999, and Blau & Kahn 2000, for a discussion). In spite of these difficulties, Altonji and Blank (1999) concluded that the extensive evidence for persistent unexplained gaps suggests that there is ongoing gender discrimination in the labour market.

15

SOCIAL CLOSURE AND GENDERED LABOUR PROCESSES Finally, two other theoretical approaches that contribute to our understanding of gender segregation will be discussed. These are gendered social closure processes (e.g., Reskin 1988; Tilly 1998; Tomaskovic-Devey 1993) and gendered organizational processes (Acker 1990). The reason for focusing on these two approaches here is that Tomaskovic-Devey and Skaggs (2002) suggested that social closure and gendered labour processes may constitute part of the explanation for women having less access to on-the-job training and to jobs with long training periods (cf., Paper IV).

Reskin (1988) argued that the basic source of the gender wage gap is not sex segregation but “…men’s desire to preserve their advantaged position and their ability to do so by establishing rules to distribute valued resources in their favour” (Reskin 1988:61).16 Hence, according to the social closure approach, the dominant group, i.e. men, will collectively use their dominance to advance their own position vis-à-vis the dominated group, i.e. women. If this is not enough, they will use their privileged position to rewrite the rules for rewards in their favour. Reskin largely built her argument on an example used by Lieberson (1985) in his critique of causal analysis. Lieberson (1985) argued that reducing educational difference between blacks and whites would not reduce the black-white income gap because whites, by being the dominant group, would find another way to maintain their advantage. In dominance systems, differentiation is the fundamental process. Gender differentiation can either be obtained through physical segregation or behavioural differentiation (i.e., task differentiation and social differentiation). Reskin put forward, among other things, evidence of social closure processes by pointing to male resistance to women’s entry into college dining clubs, private professional clubs and the Rotary (see Reskin 1988 for references).17

Tomaskovic-Devey and Skaggs (2002) argued that on-the-job training may be one of the most powerful social closure mechanisms through which the male workforce can exclude women from desirable jobs. This is, of course, particularly likely with respect to informal on-the-job training, that is, when an employee receives instructions or help from a fellow worker or supervisor. However, the extent to which male workers, as an advantaged group, can exclude women from formal on-the-job training18 is conditioned by their ability

16

to take part in decisions regarding the allocation and implementation of such training.

Acker (1990) argued that organizations are inherently gendered in fundamental ways. By this she means that “…advantage and disadvantage, exploitation and control, action and emotion, meaning and identity, are patterned through and in terms of a distinction between male and female, masculine and feminine” (Acker 1990:146). An illuminating example of this is the abstract definition of a job or a hierarchy. In organizational logic, the abstract job is filled by a disembodied worker who exists only for the work. Hence no factors outside the job – such as for instance children – can impinge upon the job (Acker 1990). Given that women are the ones that take prime responsibility for children the concept of ‘a job’ is highly gendered even though organizational logic presents it as gender neutral. Hierarchies are also gendered to the extent that they sort workers, i.e. women and men, into hierarchical positions on the basis of ‘commitment’. Those who are seen as fully committed to their job are more suited to responsibility and authority than are those who must divide their commitment (for instance between work and childcare). Gendered organizational processes not only affect processes within the organization, but the division between paid and unpaid work is also partly created through organizational processes. As a consequence, some aspects of gender identity too are products of organizational processes and pressures. Hence, widely disseminated cultural images of gender are invented and reproduced in organizations, partly as a consequence of the disembodied abstraction of a job, the job valuation process, hierarchies, etc.

As is evident from the above, men and women are likely to end up in jobs at different hierarchical levels due to gendered organizational structures and practices. With regard to on-the-job training that leads to promotions, men would, according to this approach, be more likely to be assigned to this training than women. In the fourth paper of this thesis, gender differences in formal on-the-job training, as well as differences in types of training, are studied. Gendered organizational processes, together with gender discrimination and especially prejudiced discrimination, are suggested as possible mechanisms through which these differences can come about. In the following section, the four papers are summarized, starting with those focussed on gender differences in the household.

17

PAPER I: The Reproduction of Gender. Housework and Attitudes Towards Gender

Equality in the Home Among Swedish Boys and Girls

The housework Swedish girls and boys age 10 to 18 do, and their attitudes towards gender equality in the home are studied. One aim is to see whether the work children do is gendered and if so, whether they follow their parents’, often gendered, pattern in housework. A second aim is to see whether parents’ division of work and/or children’s own work is related to the children’s attitude to gender equality in the home. The data used are the Swedish Child Level of Living Survey 2000. This survey, in combination with the Level of Living Survey 2000, gives us detailed first-hand information both from parents and from children on housework, attitudes to gender equality in the home, and a number of other factors, which enables analyses of parents’ and children’s housework that are rare in an international perspective.

Results indicate that girls and boys in two-parent families are more prone to engage in gender atypical work the more their parent of the same sex engages in this kind of work. The fact that girls still do more housework than boys in all families, independent of the parental division of housework, the mother’s educational level and her work hours, indicates that housework to some extent signifies gender also to children. However, the more both boys and girls help out in the household, the more likely they are to state that gender equality in the home is important. No relation between parents’ division of work and the child’s attitude towards gender equality in the home is found.

PAPER II: Dependence Within Families and the Household Division of Labour –

A Comparison Between Sweden and the United States.

- together with Magnus Nermo. Submitted.

This paper assesses the relative explanatory value of the resource-bargaining perspective and the doing gender approach for the division of housework in the United States and Sweden, in the period from the early 1970s to 2000. By comparing data from the Swedish Level of Living Survey with the American Panel Study of Income Dynamics, we conclude that housework is truly gendered

18

work in the US as well as in Sweden during the entire period. However, an empirical analysis of the doing gender approach – as tested by Brines (1994) and others (Bittman et al. 2003; Greenstein 2000) – indicates that gender-creating processes prevail more in the US than in Sweden. The results indicate that women in the United States, but not Sweden, use time spent in housework as a way to neutralize gender deviance. Our results question the finding from Brines (1994). She found that only dependent men took part in gender display, i.e. did less housework than would be predicted by bargaining theories. We, on the other hand, find indications that women married to or cohabiting with dependent men increase their housework, as if to shore up their husbands’ masculinity. This leads to a curvilinear effect of women’s relative earnings on housework in three out of four years in the US: in 1981, 1991 and 1999.

PAPER III Divorce and Labour-Market Outcomes. Do Women Suffer or Gain?

Published in Jonsson, J. O. and C. Mills. 2001. “Cradle to Grave. Life-course change in modern Sweden.” Durham, UK: Sociologypress.

Earlier studies indicate that women in Sweden tend to increase their working hours in paid labour after a separation, partly because the negative economic consequences of a divorce often are considerable. Can changes like these and others partly derived from separation/divorce influence individuals’ opportunities in the labour market?

The data used in this study are retrospective work and family histories collected in the Swedish Level of Living Survey in 1991. These data render possible complex analyses of the interconnected nature of family and work. A hazard regression model with competing risks reveals that women’s chances of improving their occupational prestige appear to be better after a divorce compared to before; that is, the first job change after a divorce is often to the better for women. Increased working hours and perhaps also increased energy invested in the job may pay off in better occupational opportunities. Worth noting, however, is that the outcome among women with a less firm labour market attachment is more often to a job of lower prestige than to a job of higher prestige. Hence, after a divorce, the labour market outcome for women is to some extent conditioned by their labour market attachment at the time of

19

divorce. Men, on the other hand, in most cases seem to suffer occupationally from divorce. For separated men the risk of negative changes in occupational prestige is greater than for cohabiting men.

PAPER IV Formal On-the-job Training.

A Gender-typed Experience and Wage-related Advantage?

Published in European Sociological Review, Vol. 20, No. 1, January 2004.

Formal on-the-job training (FOJT) can have a positive impact on wages and on promotion opportunities. According to theory and earlier research, a two-step model of gender inequality in FOJT is predicted: First, women are less likely than men to take part in FOJT and, second, once women do get the more remunerative training – such as general training and training that increases promotion opportunities – they are not rewarded for their new skills to the same extent as men are. The ability to distinguish between different types of FOJT, such as firm-specific training, industry-specific training and general training together with training that increases promotion opportunities – a type of training that cuts across the other three – is important, not least with regard to estimating the rewards of training. However, few earlier studies have been able to make these distinctions.

Pooled cross-sectional data from the Swedish Survey of Living Conditions in the mid-nineties were used. Logistic and OLS regression models were estimated to address the hypotheses. Results show that women are significantly less likely than men to take part in FOJT. Among those who do receive training, women are more likely to take part in industry-specific training, whereas men are more likely to participate in general training and training that increases promotion opportunities. The two latter forms of training significantly raise a man’s annual earnings but not a woman’s. Hence, the predicted model is supported and it is argued that this gender inequality is partly due to employers’ discriminatory practices.

20

REFERENCES Acker, J. (1990). Hierarchies, Jobs, Bodies: A Theory of Gendered

Organizations. Gender and Society, 4, 139–158. Ahrne, G., & Roman, C. (1997). Hemmet, barnen och makten. Förhandlingar

om arbete och pengar i familjen. [The home, the children and the power. Negotiations concerning work and money in the family]. SOU 1997:139. Stockholm: Fritzes.

Altonji, J. G., & Blank, R. M. (1999). Race and Gender in the Labor Market. In Ashenfelter, O. C., & Card, D. Handbook of Labor Economics. Volume 3C. Amsterdam: Elsevier Science.

Andersson, G. (2002). Children’s experience of family disruption and family formation: Evidence from 16 FFS countries. Demographic Research, 7, 343–363.

Andersson, G. (2003). Dissolutions of unions in Europe: A comparative overview. MPIDR Working Paper 2003-004. Rostock: Max Planck Institute for Demographic Research.

Axelsson, C. (1992). Hemmafrun som försvann. Övergången till lönearbete bland gifta kvinnor I Sverige 1968-81. [The housewife who disappeared. The transition to wage work among married women in Sweden 1968-81]. Stockholm: Institutet för social forskning.

Baude, A. (1992). Kvinnans plats på jobbet. [The woman’s place at work] Stockholm: SNS Förlag.

Becker, G. S. (1971). The Economics of Discrimination. Second Edition. Chicago: University of Chicago Press.

Becker, G. S. (1975). Human capital: A Theoretical and Empirical Analysis with Special Reference to Education. Second Edition. New York: National Bureau of Economic Research.

Becker, G. S. (1981). A Treatise of the Family. Cambridge: Harvard University Press.

Becker, G. S. (1985). Human Capital, Effort, and the Sexual Division of Labor. Journal of Labor Economics, 3, 33–58.

Bittman M., England, P., Sayer, L., Folbre, N., & Matheson, G. (2003). When Does Gender Trump Money? Bargaining and Time in Household Work. American Journal of Sociology, 109, 186–214.

Blau F. D., & Ferber, M. A. (1992). The Economics of Women, Men, and Work. Englewood Cliffs, NJ: Prentice-Hall.

21

Blau, F. D., & Kahn, L. M. (1996). Wage Structure and Gender Earnings Differentials: An International Comparison. Economica 63 (Supplement) S29–S62.

Blau, F. D., & Kahn, L. M. (2000). Gender Differences in Pay. Journal of Economic Perspectives, 14, 75–99.

Blood, R. O., & Wolfe, D. M. (1960). Husbands and Wives. Glencoe, IL: Free Press.

Brines, J. (1994). Economic Dependency, Gender, and the Division of Labor at Home. American Journal of Sociology, 100, 652–688.

Bygren, M. (2001). Career Outcomes in the Swedish Labor Market. Swedish Institute for Social Research – Dissertation Series no 51. Stockholm: Swedish Institute for Social Research.

Båvner, P. (2001). Half full or Half Empty? Part-time Work and Well-being Among Swedish Women. Swedish Institute for Social Research – Dissertations Series no. 49. Stockholm: Swedish Institute for Social Research.

Casper, L. M., McLanahan, S. S., & Gaarfinkel, I. (1994). The Gender-Poverty Gap: What We Can Learn from Other Countries. American Sociological Review, 59, 594–605.

Charles, M. (1992). Cross-National Variation in Occupational Sex Segregation. American Sociological Review, 57, 483–502.

Christopher, K., England, P., Smeeding, T. M., & Ross Phillips, K. (2002). The Gender Gap in Poverty in Modern Nations: Single Motherhood, the Market, and the State. Sociological Perspectives, 45, 219–242.

Connell, R. W. (1985). Theorizing gender. Sociology, 19, 260–272. Elvander, N. (1974). Skattepolitik 1945-1970. En studie i partiers och

organisationers funktioner. [Fiscal policies 1945-1970. A study of the impact of parties and organizations] Stockholm: Rabén och Sjögren.

England, P. (1992). Comparable Worth. Theories and Evidence. New York: Aldine de Gruyter.

England, P. (1993). The Separative Self: Androcentric Bias in Neoclassical Assumptions. In Ferber, M. A., & Nelson, J. A. (Eds.), Beyond Economic Man. Feminist Theory and Economics. Chicago: The University of Chicago Press.

22

England, P., & Budig, M. (1998). Gary Becker on the Family: His Genius, Impact and Blind Spots. In Clawson, D. (Ed.), Required Reading: Sociology’s Most Influential Books. Amherst: University of Massachusetts Press.

England, P., & Kilbourne, B. S. (1990). Markets, Marriages, and Other Mates: The Problem of Power. In Friedland, R., & Robertsson, A. F. (Eds.), Beyond the Marketplace. Rethinking Economy and Society. New York: Aldine de Gruyter.

Fagrell, B. (2000). De små konstruktörerna. Flickor och pojkar om kvinnligt och manligt i relation till kropp, idrott, familj och arbete. [The young constructors. Girls and boys on femininity and masculinity in relation to body, sports, family and work]. Stockholm: HLS Förlag.

Fenstermaker Berk, S. (1985). The Gender Factory. The Apportionment of Work in American Households. New York: Plenum Press.

Feree, M. M. (1990). Beyond Separate Spheres: Feminism and Family Research. Journal of Marriage and the Family, 52, 866–884.

Flood L., & Gråsjö, U. (1997). Tid för barn, tid för arbete. En undersökning av svenska hushålls tidsanvändning. [Time for children, time for work. An examination of time use in Swedish households]. In Ahrne, G., & Persson, I. (Eds.), Familj, makt och jämställdhet. SOU 1997:138.

Fritzell, J., & Lundberg, O. (1994). Levnadsvillkor i förändring. [Living conditions in a state of flux]. In Fritzell, J., & Lundberg, O. (Eds.), Vardagens Villkor. Levnadsförhållanden i Sverige under tre decennier. Stockholm: Brombergs.

le Grand, C. (1997). Occupational Segregation and the Gender Gap in Sweden. Working Paper Nr. 50, Sociologiska Institutionen, Stockholms universitet.

le Grand, C., Szulkin, R., & Tåhlin, M. (2001). Lönestrukturens förändring i Sverige. [Change in the wage structure in Sweden]. In Fritzell, J., Gähler, M., & Lundberg, O. (Eds.), Välfärd och arbete i arbetslöshetens årtionde. SOU 2001:53. Stockholm: Fritzes.

Government Bill 1973:47. Förbättrade familjeförmåner inom den allmänna försäkringen m.m. [Improved family benefits in the general insurance, etc.]. Stockholm.

Greenstein, T. N. (2000). Economic Dependence, Gender, and the Division of Labor in the Home: A Replication and Extension. Journal of Marriage and the Family, 62, 322–335.

23

Gähler, M. (1998). Life After Divorce. Economic, Social and Psychological Well-being Among Swedish Adults and Children Following Family Dissolution. Swedish Institute for Social Research – Dissertations Series no 32. Stockholm: Swedish Institute for Social Research.

Gähler, M. (2001). Bara en mor – ensamstående mödrars ekonomiska levnadsvillkor i 1990-talets Sverige. [Only a mother – single mothers’ economic living conditions in Sweden in the 1990s]. In Bergmark, Å. (Ed.), Ofärd i Välfärden. SOU 2001:54. Stockholm: Fritzes.

Haddad, L., & Hoddinott, J. (1994). Women’s Income and Boy-Girl Anthropometric Status in the Côte d’Ivoire. World Development, 22, 543–553.

Henz, U., & Jonsson, J. O. (2003). Union Disruption in Sweden. Does Economic Dependency Inhibit Separation? International Journal of Sociology, 33, 3–39.

Hirschman, A. O. (1970). Exit, Voice and Loyalty. Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press.

Hobson, B. (1990). No Exit, No Voice: Women’s Economic Dependency and the Welfare State. Acta Sociologica, 33, 235–250.

Hobson, B. (2003). Making Men into Fathers: Policy Efforts to Overcome Inequalities in the Family. Paper presented at the Euresco Conference, Helsinki, September 20–21.

Hobson, B., & Takahashi, M. (1997). The Parent-Worker Model: Lone Mothers in Sweden. In Lewis, J. (Ed.), Lone Mothers in European Welfare Regimes. London: Jessica Kingsley Publishers.

Hoem, B. (1995). The Way to the Gender Segregated Swedish Labour Market. In Oppenhiem Mason, K., & Jensen, A-M. (Eds.), Gender and Family Change in Industrialized Countries. Oxford: Clarendon Press.

Hultin, M. (1998). Gender Differences in Workplace Authority: Discrimination and the Role of Organizational Leaders. Acta Sociologica, 41, 99–113.

Hultin, M. (2003). Some take the glass-escalator, some hit the glass ceiling? Career Consequences of Occupational Sex Segregation. Work and Occupations, 30, 30–61.

Hultin, M., & Szulkin, R. (1999). Wages and Unequal Access to Organizational Power: An Empirical Test of Gender Discrimination. Administrative Science Quarterly, 44, 453–472.

24

Human Development Report (1995). New York: Oxford University Press for the United Nations Development Programme (UNDP).

Human Development Report (2000). New York: Oxford University Press for the United Nations Development Programme (UNDP).

Hörnqvist, M. (1997). Familjeliv och arbetsmarknad för män och kvinnor. [Family life and labour market for men and women]. In Ahrne, G., & Persson, I. (Eds.), Familj, makt och jämställdhet. SOU 1997:138.

Jonsson, J. O. (1999). Explaining Sex Differences in Educational Choice. An Empirical Assessment of a Rational Choice Model. European Sociological Review, 15, 391–404.

Jonsson, J. O., & Mills, C. (2001). Giving birth without giving up: return to employment and return to work amongst women. In Jonsson, J. O., & Mills, C. (Eds.), Cradle to Grave. Life-course change in modern Sweden. Durham: Sociologypress.

Kanter, R. M. (1977). Men and Women of the Corporation. New York: Basic Books.

Klevmarken, N. A., & Olovsson, P. (1993). Household, market and non-market activities. Stockholm: Industrial Institute for Economic and Social Research.

Lazear E. P., & Rosen, S. (1990). Male-Female Wage Differentials in Job Ladders. Journal of Labor Economics, 8, 106–123.

Lieberson, S. (1985). Making it Count. Berkeley: University of California Press. Lopata, H., & Thorne, B. (1978). On the term ‘Sex Roles’. Signs, 3, 718–721. Lorber, J. (1994). Paradoxes of Gender. New Haven: Yale University Press. Lundberg, S., & Pollak, R. (1996). Bargaining and Distribution in Marriage.

Journal of Economic Perspectives, 10, 139–158. Lundh, C., & Olsson, R. (1994). Från arbetskraftsimport till flyktinginvandring.

[From importation of manpower to refugee immigration]. Stockholm: SNS förlag.

Manser, M., & Brown, M. (1980). Marriage and Household Decision-Making: A Bargaining Analysis. International Economic Review, 21, 31–44.

Marini, M. M., & Fan, P-L. (1997). The Gender Gap in Earnings at Career Entry. American Sociological Review, 6, 588–604.

McElroy, M. B., & Horney, M. J. (1981). Nash-Bargained Household Decisions: Toward a Generalization of the Theory of Demand. International Economic Review, 22, 333–349.

25

The National Swedish Social Insurance Board (2002). Spelade pappamånaden någon roll? – Pappornas uttag av föräldrapenning. RFV (Riksförsäkringsverket) Analyserar 2002:14.

The National Swedish Social Insurance Board (2003). RFV (Riksförsäkringsverket) Internet page: http://www.rfv.se/forsak/famba/fp/index.htm (November 2003).

Nermo, M. (1994). Den ofullbordade jämställdheten. [The incomplete equality]. In Fritzell, J., & Lundberg, O. (Eds.), Vardagens Villkor. Levnadsförhållanden i Sverige under tre decennier. Stockholm: Brombergs.

Nermo, M. (1996). Occupational Sex Segregation in Sweden 1968-1991. Work and Occupations, 23, 319–322.

Nermo, M. (1999). Structured by Gender. Patterns of sex segregation in the Swedish labour market. Historical and cross-national comparisons. Swedish Institute for Social Research – Dissertations Series no 41. Stockholm: Swedish Institute for Social Research.

Nermo, M. (2000). Models of Cross-national Variation in Occupational Sex Segregation. European Societies, 2, 295–333.

Okamoto, D., & England, P. (1999). Is There a Supply Side to Occupational Sex Segregation? Sociological Perspectives, 42, 557–582.

Orloff, A. S. (1993). Gender and the Social Rights of Citizenship. American Sociological Review, 58, 303–328.

Padavic, I., & Reskin, B. (2002). Women and Men at Work. London: Pine Forge Press.

Reskin, B. (1988). Bringing the Men Back In: Sex Differentiation and the Devaluation of Women’s Work. Gender & Society, 2, 58–81.

Reskin, B. (2003). Including Mechanisms in Our Models of Ascriptive Inequality. American Sociological Review, 68, 1–21

Rønsen, M., & Sundström, M. (1996). Maternal employment in Scandinavia: A comparison of the after-birth employment activity of Norwegian and Swedish women. Journal of Population Economics, 9, 267–285.

Statistics Sweden (1990). Familjebildning och familjeupplösning under 1980-talet. [Family formation and dissolution during the 1980s]. Demografiska rapporter 1990:1. Stockholm: Statistics Sweden.

26

Statistics Sweden (1991). På skilda banor. Så väljer kvinnor och män utbildning och yrke. [On different tracks. Choices of education and occupation among women and men]. Information om arbetsmarknaden 1991:5. Stockholm: Statistics Sweden.

Statistics Sweden (2002). På tal om kvinnor och män. Lathund om jämställdhet. [Talking about women and men]. Stockholm: Statistics Sweden.

Statistics Sweden (2003). Tid för vardagsliv. Kvinnor och mäns tidsanvändning 1990/91 och 2000/01. [Time for everyday life. Women’s and men’s time use 1990/91 and 2000/01]. Stockholm: Statistics Sweden.

Statistical Yearbook of Sweden. (2003). Örebro: Statistics Sweden. SOU 2003:16. Mansdominans i förändring. Om ledningsgrupper och styrelser.

Betänkande av Utredningen om kvinnor på ledande poster i arbetslivet. [Male dominance in a state of flux. On managerial groups and boards]. Stockholm: Fritzes.

Sundström, M., & Duvander, A-Z. E. (2002). Gender Division of Childcare and the Sharing of Parental Leave among New Parents in Sweden. European Sociological Review, 18, 433–447.

Svensson, L. (2003). Explaining Equalization. Political Institutions, Market Forces, and Reduction of the Gender Wage Gap in Sweden, 1920-95. Social Science History, 27, 371–395.

Sørensen, A. (2001). Gender equality in earnings at work and at home. In Kautto, M., Fritzell, J., Hvinden, B., Kvist, J., & Uusitalo, H. (Eds.), Nordic Welfare States in the European Context. London: Routledge.

Sørensen, A., & MacLanahan, S. (1987). Married Women’s Economic Dependency, 1940-1980. American Journal of Sociology, 93, 659–687.

Thomas, D. (1990). Intra-Household Resource Allocation: An Inferential Approach. Journal of Human Resources, 25, 635–664.

Thomas, D. (1994). Like Father, Like Son: Like Mother, Like Daughter: Parental Resources and Child Height. Journal of Human Resources, 29, 950–988.

Thorne, B. (2002). In “Symposium on West and Fenstermaker’s ‘Doing Difference’” (with Hill Collins, P., Maldonado, L. A., Takagi, D. Y., Weber, L., & Winant, H.). In Fenstermaker, S., & West, C. (Eds.) Doing Gender, Doing Difference. Inequality, Power and Institutional Change. New York: Routledge.

Tilly, C. (1998). Durable Inequality. Berkeley: University of California Press.

27

Tomaskovic-Devey, D. (1993). Gender and Racial Inequality at Work: The Sources and Consequences of Job Segregation. Ithaca, NY: ILR Press.

Tomaskovic-Devey, D., & Skaggs, S. (2002). Sex Segregation, Labor Process Organization, and Gender Earnings Inequality. American Journal of Sociology, 108, 102–128.

Wadsby, M., & Svedin, C-G. (1993). Skilsmässa – bakgrund, orsaker och följder. [Divorce – background, causes and consequences]. In Agell, A., Arve-Parès, B., & Björnberg, U. (Eds.), Modernt Familjeliv och Familjeseparationer. Stockholm: Swedish Council for Social Research.

Wernersson, I. (1977). Könsdifferentiering i grundskolan. [Gender differentiation in the compulsory school]. Göteborg Studies in Educational Sciences 22. Göteborg: Acta Universitatis Gothoburgensis.

West, C., & Zimmerman, D. H. (1987). Doing Gender. Gender & Society, 1, 125–151.

28

29

NOTES

1 The GDI is the Human Development Index (HDI) adjusted for gender inequality. The HDI measures a country’s achievements in three aspects of human development: longevity, knowledge, and a decent standard of living. Longevity is measured by life expectancy at birth; knowledge is measured by a combination of the adult literacy rate and the combined gross primary, secondary, and tertiary enrolment ratio; and standard of living is measured by GDP per capita (Human Development Report 2000).

2 They, among other things, feared that ghettos of – partly unemployed – foreigners were being established in the larger cities.

3 Paid maternity leave was introduced in 1955. To start with, women were granted three months of paid leave. In 1963, the entitlement was extended to six months (Jonsson & Mills 2001).

4 Examples of large predominantly female occupations are assistant nurses, child minders, cleaners, secretaries and teachers in pre-school and the nine-year compulsory school. Examples of male occupations are truck-drivers, construction workers, systems engineers, and mechanical engineers (Statistics Sweden 2003).

5 Note, however, that this figure is affected by the degree of female labour force participation.

6 Padavic and Reskin (2002) found that, among the OECD countries, the Scandinavian countries have some of the most segregated labour markets.

7 The comparative advantage is calculated as the ratio between the pupil’s marks in one subject and their marks in another.

8 See also Bygren 2001 who finds that women in male dominated workplaces to a higher extent are excluded from these workplaces (i.e. their work episodes more often end in unemployment).

9 According to the definition used here, an individual is considered to be in poverty if he or she lives in a household with a (size-adjusted) disposable monetary income that is less than half the median for households in the nation. The gender poverty ratio is the ratio of women’s poverty rate to men’s poverty rate.

10 As a comparison, the poverty rate among single mothers in the US is 47 percent and in the UK 32 percent (Christopher et al. 2002).

11 In these families, both spouses generally work full-time. It could however be that women with lower income work part-time because they are more family oriented than others (Henz & Jonsson 2003).

12 Flood and Gråsjö (1997) studies housework among individuals 18 to 40 years of age. They use the 1993 HUS survey (Household Market and Non Market Activities, Klevmarken & Olovsson 1993). Hörnqvist (1997) and Nermo (1994) use the 1991 Swedish Level of Living Survey (see Fritzell & Lundberg, 1994).

30

13 When single individuals with no children 20-59 years of age are studied, men on average

spend 1,9 hours per week on housework compared to 2,2 hours for women (calculations from Hörnqvist 1997).

14 Collective models of discrimination can be compared to what England (1992) refers to as monopoly or monopsony models of discrimination.

15 The outcome, however, would be the same independent of who has the taste if employers need to consider co-workers' and customers’ tastes when making hiring decisions and promotions.

16 As evidence of this conclusion, Reskin pointed to the lack of a clear link between changes in the sex segregation index and changes in the gender wage gap in the US.

17 Although Reskin still adheres to the importance of social closure processes for occupational sex segregation, she regrets that we even today know little about “…how specific workplace mechanisms favour members of dominant groups to varying degrees, and how extra-workplace factors lead organisations to alter or maintain those rules.” (Reskin 2003: 4).

18 Formal on-the-job training is often characterized by employer-run training sessions.