Gender, Work and Space

Susan Hanson, Geraldine Pratt

• Publisher: Routledge
• Year: 2003
• eBook ISBN: 9781134857609
• Edition: 1
• Topic: Physical Sciences
• Subtopic: Geography
• Index: Geography

Gender, Work and Space explores how social boundaries are constructed between women and men, and among women living in different places. Focusing on work, the segregation of men and women into different occupations, and variations in women's work experiences in different parts of the city, the authors argue that these differences are grounded, constituted in and through, space, place, and situated social networks.
The sheer range and depth of this extraordinary study throws new light on the construction of social, geographic, economic, and symbolic boundaries in ordinary lives.

1
SPATIAL STORIES AND GENDERED PRACTICES

Every story is a travel story—a spatial practice.

Michel de Certeau, The Practice of Everyday Life, p. 115

We intend to tell spatial stories about the construction of gender differences. We write about lives in contemporary Worcester, Massachusetts and structure the plot around women’s and men’s labor market experiences. Our interest in paid employment does, however, take us well beyond the workplace, into neighborhoods and homes. Our argument is that social and economic geographies are the media through which the segregation of large numbers of women into poorly paid jobs is produced and reproduced.

This book is also a travel story, written about gender, place, and space, through time. We began this research in 1986, and although we attempt to draw together the segments that have been published over the last six years and reinterpret them as a whole, we would misrepresent the process involved, as well as the real movement that has occurred within feminist geography, if we were to suppress theoretical shifts and conceptual reorientations. Over time, we have chosen to address different theoretical literatures, which take varying questions as being important, and have different objectives and varying styles of representation. We see this theoretical restlessness as having positive consequences. Each theoretical tradition offers a different perspective on women’s and men’s lives and can be used to highlight the underlying assumptions and silences of the others. Each addresses a different audience; by speaking in multiple conceptual languages, we hope to reach different communities of readers, not just sympathetic feminist geographers, but also political economists, more orthodox economists and sociologists, and wider geographical and feminist audiences.

Perhaps most important, each theoretical tradition draws one into different spatial stories. We borrow the term “spatial stories” from de Certeau (1984), who distinguishes among different ways of representing geography, and places a higher value on some forms than on others. He uses the experience of gazing from the summit of the World Trade Center in New York City as a metaphor for a detached, scientific, scopic representation of place and compares it unfavorably to the experience of walking through the city. He privileges the description of the “tour” as a less abstracted representation of geographical experience than that offered by the map. De Certeau’s representational distinctions have been used to good effect in critiques of detached geographical accounts (Deutsche 1991; Gregory 1990). We are reluctant, however, to subscribe to de Certeau’s hierarchies of difference; instead we appreciate the utility of both gazing from the skyscraper and walking through the street, with the understanding (which absorbs de Certeau’s critique of scopic representation) that both perspectives are partial.

Different theoretical traditions encourage us to think from different vantage points. The literature on occupational segregation is, for example, largely quantitative and analysts search for generalizations across all women. Feminist cultural criticism, in contrast, often starts from the autobiographical, situated account. Geographical literature on labor markets typically gives priority to economic and structural determinants over the cultural. Different literatures start from different presuppositions and assume different points of closure. We follow Flax (1990) and Ferguson (1993) in seeing the advantage of working with multiple theoretical traditions: in their differences and complementarities they force a richer set of spatial stories. We make no claims to exhaustion; our encounters with multiple theoretical and empirical traditions remind us of the partiality of our narratives.

In this chapter, we provide an overview of these theoretical traditions, to show how each contributes to an understanding of the spatial construction of gender relations. We deal with them more or less in the order with which they have influenced our thinking, beginning with the economic, sociological, and feminist geographical literatures on occupational segregation, moving to feminist and industrial geographers’ treatments of occupational segregation and labor market segmentation, and finally to theorizing about gender identity and space within contemporary feminist theory.

OCCUPATIONAL SEGREGATION—WHAT’S SPACE GOT TO DO WITH IT?

It is unexceptional for women in most parts of the world to hold jobs outside of the home, even when we have the responsibility of mothering very young children. In the U.S. in 1990, 57 percent of women between the ages of 16 and 64 were in the waged labor force, as were 59 percent of mothers of children under age 6. Women constitute a significant proportion (45 percent in 1990) of all waged workers (U.S. Bureau of the Census 1990). Similarly, in 1990, 58.6 percent of married women in the U.K. and 57.9 percent of married women in Canada were in the paid work force (McDowell 1991a; Statistics Canada 1993).

Women do, however, continue to participate in the labor force in ways that differ from men. One point of difference is that women as a group tend to earn less. At every educational level, women in the U.S. earned less in 1990 than men with the same amount of schooling, and although the pay gap between women and men who work full time has declined in recent years, it continues to widen with age. In 1991 college-educated women between the ages of 25 and 34 earned on average 90 cents for every dollar earned by a man of similar age and education, an increase of five cents on the dollar since 1981 (Steinbruner and Medoff 1994). The 1991 wages of college-educated women between 45 and 54 years of age, however, were only 62 cents for every dollar earned by men of the same age and education, the same amount earned by male high school graduates (Medoff 1994; for comparable 1986 data, see Reskin and Roos 1990:20).

Some may interpret these data as good news, evidence that gender inequality is declining over time and is especially on the wane among younger age cohorts. The compression of the gender wage gap is due in part, however, to a decline in men’s wages as well as to the increase in women’s (Steinbruner and Medoff 1994). And the wage differentials between the age groups reflects the fact that most women enjoy rather less in the way of wage increases over their lifetimes, than do most men (Rosenfeld 1979). The enduring gender wage gap reflects not only different rates of career mobility within particular occupations but also the fact that women and men tend to work in different occupations and industrial sectors. For example, in 1980, 48.9 percent of employed American women (compared to 16.1 percent of men) worked in administrative support (including clerical) and service occupations, while 28.5 percent of all men and 3.3 percent of women worked in precision production and craft, and transportation occupations (Reskin and Roos 1990:4). Typically the sectors and occupations filled by women tend to have lower income ceilings, poorer benefits, and less career mobility.

The tendency for women to have different occupations from men is what is referred to as occupational segregation; it is a phenomenon that appears to be remarkably persistent. Historical studies (Gross 1968; Beller 1984; Jacobs 1989) indicate that levels of occupational segregation in the United States were essentially constant from 1900 to 1960. During the 1970s sex-based occupational segregation did decline slightly; even so, in 1980 almost half of all American women and 53 percent of men worked in occupations that employed, respectively, at least 80 percent women and men (Rytina and Bianchi 1984). Reskin and Roos (1990) have made a close study of the 33 male-dominated occupations into which American women seemed to have made significant inroads during the 1970s, increasing their representation by at least 9 percent. On closer examination, they find very little evidence for genuine integration. Women are either ghettoized into low-status specialties, low-paying industries, and part-time work or the occupation has been re-segregated as predominantly female (e.g., this is the case for insurance adjusters, examiners, and investigators). “The answer to our question of whether occupation-level desegregation translated into job-level integration is a resounding no. Women and men tend not to work alongside one another on the same jobs in these newly desegregating occupations’ (Reskin and Roos 1990:72).

Opinions about whether occupations should be reconstituted so that they are filled equally by women and men, and how this might be accomplished, depend on one’s analysis of the sources of occupational segregation. Much of the debate about the reasons for continuing occupational segregation can be read as being implicitly geographical, insofar as the argument turns around where the causes are located: in the home and within the family or through workplace-based practices.

Occupational segregation begins at home

A range of theorists locate occupational segregation within the home, although they vary quite radically in their normative and political stances. Those who explain occupational segregation and gender-based wage differentials in terms of human capital theory (e.g., Mincer and Polachek 1978; Mincer and Ofek 1982; Polachek 1981) tend not to see either of these labor market outcomes as being particularly problematic (or at least, not as signaling a structural problem within the labor market). They locate women’s labor market position in terms of their household responsibilities and interpret a woman’s decision to work in a female-dominated occupation as a rational choice, made within the nexus of these additional work obligations. The rationale for this decision hinges on a woman’s expectation of moving into and out of the labor force as the demands of childbearing and child care ebb and flow. According to this line of reasoning, a woman’s expectation that she will spend fewer years in the labor force over the course of her life than will a man makes it irrational for her to spend many years in pre-job training. Additionally, human capital theorists argue that traditionally female occupations are less likely than are male occupations to have apprentice-type training periods built into the early years on the job, during which time wage levels are suppressed. They argue that female occupations thus have lower “start-up” costs, an attractive feature to an individual who intends to move in and out of the labor force.

Human capital theory has received many challenges, both empirical and theoretical, which we agree with and review in Chapter 5, but it should be recognized that some feminist attempts to explain occupational segregation also locate its origins within the home, in women’s heavy domestic workloads (without assuming the rationality of individual choices or drawing the same normative conclusions). This was particularly true of feminist accounts in the late-1970s and early-1980s. Participants in the domestic labor debate, for example, interpreted women’s position almost exclusively in terms of their domestic responsibilities, seeing women’s provision of this labor as being functional to capitalists insofar as it cheapens the costs of social reproduction and indirectly dampens wage demands. The Marxist and socialist feminists who were engaged in this debate disagreed about whether the burden of domestic work was thrust upon women by capitalists or by men as patriarchs, that is, they disagreed about the relationship between capitalism and patriarchy, but there was general agreement that subordination within the home in large part explained women’s subordination in the labor force. (For a review of these arguments, see Tong 1989.) It is worth noting, however, that domestic labor theorists largely draw out the temporal rather than the spatial relations between home and work, as, for example, in research that emphasizes the implications of women’s heavy domestic workloads for their types of waged employment (Hartmann 1987). Our interest in geography leads us to complement these temporal stories with spatial ones.

Occupational segregation begins at work: labor market segmentation theory

In the early-1980s many feminists attempting to explain occupational segregation turned their attention away from the household and the individual choices made in the context of familial responsibilities, toward the workplace, in line with a more structural reading of segmented labor markets (Game and Pringle 1983). From England and Farkas (1986:154) comes a strong statement that removes the process of occupational segregation from the home: “women’s domestic responsibilities cannot explain their absence from most male-dominated jobs.” Others (e.g., Reskin and Roos 1990; Walby 1986) are equally adamant that women’s family roles are largely irrelevant to their occupational choices and that the origins of occupational segregation must be sought within the labor market.

Segmented labor market theorists see the labor market as being divided into different segments, not as the result of rational choices on the part of employees and inequalities in the attainment of human capital that are structured outside of the labor market, but as the result of processes internal to the labor market itself (Morrison 1990; Peck 1989). Segmentation theorists cite a diverse range of causes for segmentation, including technological requirements, product market instability, employer discrimination, and employers’ attempts to control labor and labor costs through divisive practices. In an idealized conception, the labor market is divided into internal (or primary) and external (secondary) segments. In the internal segment, wages and job mobility are determined by institutional rules, often negotiated by employers and unions, and not determined exclusively through market forces. Workers in the internal or primary sector tend to earn better wages, enjoy more job stability and better working conditions, and have more promotional possibilities than do those in the external or secondary segment. There is little mobility across sectors, and workers in the secondary sector effectively become trapped within it. Central to segmented labor market theory is the point that workers’ labor market preferences are shaped within segments of the labor market itself, not outside of the labor market as is posited in neoclassical accounts (Lang and Dickens 1988). (For reviews of segmented labor market and human capital theories of occupational segregation, see Morrison 1990; Peck 1989; Pinch 1987; Redclift and Sinclair 1991.)

Segmented labor market theory has been criticized because of its relative inattention to the processes that lead certain categories of workers into different segments, namely white males into the primary segment and women and visible “minorities” into the secondary ones (Redclift and Sinclair 1991). A number of theorists have, however, used segmented labor market theory as a framework, into which they insert the dynamics of patriarchy at work. Explanations for women’s exclusion from the primary segment build on the sexist practices of male employers and employees. Male employers may be reluctant to hire women for the most prized jobs because of gender stereotypes, worries about complaints from male employees, and their more general fears about losing male advantage: “Dispensing with sex differentiation at any level threatens it at every level, so ignoring sex in the labor queue challenges the advantaged positions of male employers and managers” — both inside and outside the workplace (Reskin and Roos 1990:37). White, male employees also have organized through unions and professional organizations to shelter jobs for themselves (Cockburn 1983; Reskin and Roos 1990; Rubery 1978; Walby 1986).

Although the debate about how best to explain occupational segregation has been implicitly geographical in the sense of attending to where the locus of causation lies, until recently few have explored this geography or have drawn connections between relations within the home and the workplace in other than temporal terms. There has been an unproductive tendency to claim the importance of one side of the duality-either home or work. Peck remarks in reference to early segmentation models: “[t]he appreciation of the supply-side of the labor market in these models was not particularly sophisticated” (1989:45–46). Later segmentation theory, associated with the Cambridge labor market segmentation school, has placed more emphasis on labor supply (Peck 1989; Morrison 1990), but, in general, the connections between different spheres of life remain relatively undeveloped, and when connections are made, the geography of the relations between home and work tends to be ignored.

Does occupational segregation begin with networks?: economic sociology

One emerging strand of thought that has insisted on the inseparability of economic and social realms of life (though not in the context of gender-based occupational segregation) directs attention to what Granovetter (1985) calls embeddedness—the embeddedness of individuals in networks of association and the embeddedness of economic activity in social life. Referred to as economic sociology, this work speaks not of human capital, accumulated by atomized individuals, but of social and cultural “capital,” forms of noneconomic knowledge that affect the economic practices of individuals and groups as well as access to particular occupations (Granovetter 1985, 1988; Fernandez Kelly 1994). Networks of personal contacts and the information, norms, and values that are filtered through them do not respect the neat divisions scholars have drawn between the economic and the social, between production and reproduction, between work and home; they are the medium through which individuals develop aspirations and, often, acquire access to jobs.

Of distinct importance are the type and quality of resources that individuals in different networks of social relations command: Are the people in one’s social network relatively similar (to oneself and to each other), or are they diverse? What range of experiences have they had and what kind of information do they have access to? Are they well-connected to people in other networks with different characteristics? Granovetter and Tilly (1988) have argued that understanding personal networks is central to understanding labor market inequalities. They see these inequalities as resulting from bargaining and conflict among employers, workers, households, organizations, and government. In their view, personal networks and the resources they represent are crucial to the outcome of this conflict (see also Manwaring 1984; Blackburn and Mann 1979).

Although the term “embeddedness” has (to us, at least) a spatial connotation of rootedness in place, much of the discussion within economic sociology has not explicitly exploited the geographic dimensions of this concept. A recent, and fascinating, exception is Fernandez Kelly’s (1994) study of teen pregnancy and motherhood in the Baltimore ghetto, in which she explores the ways in which teenagers’ access to information and cultural resources depends on physical and social location—in this case, the neighborhood. Our Worcester study weaves together the insights of economic sociology, of labor market segmentation theory, and of geography to understand how different structures of employment opportunity are constructed in place.

Geographers have long been attentive to the spatiality of urban life, although the connections they have drawn between urban form and women’s disadvantaged position in the formal labor market have, until recently, remained either implicit or highly generalized. Early work by feminist geographers tended to focus on the friction of distance: on gender differences in travel-activity patterns and on the spatial separation of residential and employment districts. More recently, economic geographers have paid closer attention to localities and to variations in labor market segmentation across places. Each tradition tends to tell different spatial stories, ones that could inform each other in productive ways.

Commuting patterns and women’s waged labor

In the mid-1970s, feminist geographers began to make considerable effort to understand how spatial constraint enters, in a very general sense, into processes of occupational segregation (for reviews, see MacKenzie 1989; Bowlby et al. 1989). A spate of time-space geographies and journey-to-work studies undertaken in the 1970s and early 1980s documented the spatial constraints that women experience. These studies revealed that women tend to travel less frequently, over shorter distances, and via different means of travel than do men. They also explored the implications of these travel patterns for access to services and paid employment (Tivers 1985; Hanson and Hanson 1980, 1981; Forer and Kivell 1981; Madden 1981; Howe and O’Connor 1982; Pas 1984). Gradually a direct link between commuting time/distances and occupational segregation emerged from this literature with the finding that women in female-dominated occupations have shorter worktrips than other women (Hanson and Johnston 1985; Singell and Lillydahl 1986). More generally, MacKenzie and Rose (1983) argued that the spatial separation between residential suburb and urban workplaces is integral, not incidental, to the conceptual and practical separation of home and economy and to the difficulties that women experience in combining domestic and wage labor: women located in suburban residential areas are unlikely to travel the long distances required to get to urban workplaces.

Because our research emerged out of and extends this tradition, it is important to reflect on the characteristics of the spatial stories that are associated with it. Geography has tended to be read, within this strand of feminist geography, as distance and separation, highlighting the separation between home and work and the distance to services and well-paying jobs. In addition, this work focused primarily on gender differences and tended to submerge the differences that exist among women. Only relatively recently has attention been given to variations in home-employment distances among different groups of women (Johnston-Anumonwo 1992, 1994; McLafferty and Preston 1991, 1992). With access to different housing and labor submarkets, the experiences of African American and Latina women, for example, are often very different from those of white women.

This early work also paid insufficient attention to differences in the practice of occupational segregation across places and how these practices are constituted through and in place. In many cases, women’s and men’s lives have been read with very li...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• List of Plates
• List of Figures
• List of Tables
• Acknowledgments
• 1 Spatial Stories and Gendered Practices
• 2 Continuity and Change in Worcester, Massachusetts
• 3 The Worcester Study
• 4 The Friction of Distance and Gendered Geographies of Employment
• 5 Household Arrangements and the Geography of Employment
• 6 Employer Practices, Local Labor Markets, and Occupational Segregation
• 7 Communities, Work, and Gender Relations
• 8 Crossing Boundaries
• Appendix
• Notes
• References