Migrant workers, focused on Ireland

Reproducing Labor Inequalities: Challenges for Feminists Conceptualizing Care at the Intersections of Gender, Race, and Class Author(s): Mignon Duffy Source: Gender and Society, Vol. 19, No. 1 (Feb., 2005), pp. 66-82 Published by: Sage Publications, Inc. Stable URL: https://www.jstor.org/stable/30044569 Accessed: 11-04-2019 16:01 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

https://about.jstor.org/terms

Sage Publications, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Gender and Society

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

REPRODUCING LABOR INEQUALITIES

Challenges for Feminists Conceptualizing Care at the

Intersections of Gender, Race, and Class

MIGNON DUFFY

Brandeis University

The author uses census data to assess the consequences of two alternative theoretical formulations of

care work for understanding the intersections of gender race, and economic inequalities in paid care.

The nurturance conceptualization focuses on care as relationship while the reproductive labor frame-

work includes both relational and nonrelational jobs that maintain and reproduce the labor force. An

empirical application of both models to the labor market shows that placing increasing theoretical

emphasis on nurturant care privileges the experiences of white women and excludes large numbers of

very-low-wage workers from consideration.

Keywords: care work; reproductive labor; gender, race, and class inequalities

Recently, scholars and activists have focused attention on the contemporary crisis in care in the United States, particularly on the inadequacy of care resources and concomitant low wages and high turnover in care jobs. The combination of increased women's participation in the paid labor force and a dramatic aging of the population has left the United States with what Arlie Russell Hochschild (1998) has

AUTHOR' S NOTE: I would like to thank Karen Hansen and Janet Giele for their invaluable guidance and support in more ways than I can mention throughout this project. Diane Purvin read multiple ver-

sions of this article and always provided thoughtful and meaningful feedback. Many thanks to her and to

the rest of the Feminist Social Science Research Group. lam also grateful to Cameron Macdonald, Sally

Bould, AnnaAllocco, Sarah Wright, and Michael Duffy for their helpful comments on various versions of

the article and to Trent Alexander at the University of Minnesota and Grant Ritter at the Heller School

for being my statistical gurus. Finally, thank you to Dr Christine Williams and the anonymous Gender &

Society reviewers for their insightful suggestions that have made this article much stronger This research was supported in part by a Dissertation Year Fellowship from Brandeis University and the Missy Carter Doctoral Dissertation Award from the Communities, Families, & Work Program at Brandeis. An earlier version of this article was presented at the American Sociological Association Annual Meetings in Atlanta in August 2003.

REPRINT REQUESTS: Mignon Duffy, 127 Brown Avenue, #2, Roslindale, MA 02131.

GENDER & SOCIETY, Vol. 19 No. 1, February 2005 66-82

DOI: 10.1177/0891243204269499

© 2005 Sociologists for Women in Society

66

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 67

called a "care deficit." Amid stories of substandard day care centers, unhealthy con- ditions in nursing homes, and shortages of qualified nursing personnel in hospitals,

there is mounting concern about the quantity and quality of care available for chil-

dren and the elderly as well as for those who are ill or disabled. Feminist scholars have been among those making the connection between these social problems and

the low wages and poor working conditions of day care providers, health aides, and other workers who provide care for pay in the labor market. Especially concerning

to feminists is the concentration of women, particularly women of color, in these low-wage jobs.

Mona Harrington (1999, 21) warned, "We are heading towards hardening in-

equality in the creation of a new, low-wage servant class to do our caretaking for

us." She continued, "Depending on these workers, we create not just a servant class but one made up of racial and ethnic minorities." While scholars have long empha-

sized the gendered division of care work within families, the shift of focus to

include market care work has been accompanied by increased concern with racial and ethnic inequalities. A number of scholars have argued that white middle- and

upper-middle-class women have made gains in the labor force in part by transfer-

ring the care work that used to be their full-time unpaid job onto other women, often

poor women, immigrants, and women of color (Hondagneu-Sotelo 2000; Milkman 1998; Nakano Glenn 1992; Romero 1992; Tronto 2002). Mary Romero (1992, 15)

observed, "Domestic service accentuates the contradiction of race and class in fem-

inism, with privileged women of one class using the labor of another woman to escape aspects of sexism." Joan Tronto (2002) argued that feminists are to some extent accountable for the exploitation of women of color in low-wage care work through this transfer because of the feminist role in fighting for the opening of the

professions to women. It is in part this sense of accountability, along with the heightened awareness of the importance of addressing racial/ethnic and other

inequalities as part of the feminist project, that has made the issue of racialization of low-wage care a central concern for feminist scholars.

This concern was the impetus for the larger research project out of which the current article emerged. I set out to document and analyze the historical develop- ment of the paid care labor force, focusing on the gender, race, and immigration sta-

tus of the care workers as well as the wage levels of the jobs. What I soon discovered

was that the story that emerged about the racialization of low-wage care depended very much on how I defined care at the outset. Despite the recent explosion of scholarship on this topic (or perhaps because of it), there remains a lack of consis- tent conceptual clarity about what kinds of activity constitute care work. In this arti-

cle, I will present the two major theoretical directions that I identified in the care lit-

erature and explore the implications of these two distinct conceptualizations for analyzing race and class inequalities in paid care.

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

68 GENDER & SOCIETY / February 2005

CONCEPTUALIZING CARE

In a review of the care work literature, I identified two major conceptual frame-

works for understanding care, which I will call nurturance' and reproductive labor. While the universe of work and workers included in each of these definitions is

broadly overlapping, these are two distinct perspectives on the phenomenon with

different theoretical starting points. As I will show in this article, they also have quite different implications for the empirical study of care work and for the devel-

opment of conceptual and political models of care and inequalities.

In both scholarly and popular arenas, the most common understanding of care is

that of work that involves caring for children, the elderly, and those who are ill or

disabled. Both explicitly and implicitly, authors of recent calls for alarm over the state of care focus their arguments on these "dependent" populations (Harrington

1999; Heymann 2002; Stone 2000). The field remains dominated by studies of

family caregiving and paid child care workers, nurses, and home health aides. How-

ever, one of the tasks of care scholarship as a field is to move beyond commonsense

understandings of care to develop sound conceptual and theoretical frameworks.

The tendency in theoretical development has been to move away from the associa-

tion with the dependency of care receivers as the defining characteristic of care work.

Nurturance

One of the major theoretical directions in the field has been to reframe care as

relationality and interdependence. In framing care as nurturance, scholars define

care in terms of the nature of the activity itself rather than based on the population to

which it is directed. Feminist scholars have written extensively about care as a pro-

cess or practice that has a strong emotional dimension and is based on human con-

nection in relationship (Abel and Nelson 1990; Cancian and Oliker 2000; Folbre 2001; Noddings, Gordon, and Benner 1996; Russell Hochschild 1998; Tronto

1993; Tronto and Fisher 1990). While these authors do not formulate care in pre-

cisely the same way, the common thread is that they discuss care as a unique prac-

tice, skill, or way of thinking with an emphasis on relationality.

Joan Tronto and Berenice Fisher (1990), for example, described care as a pro-

cess with four "intertwining" phases, each of which is embedded in relationship

and responsiveness. "Caring about" involves paying attention to the needs of oth-

ers, a part of the process that presupposes a relational connection and includes an emotional dimension. They label as "taking care of" taking responsibility for meet-

ing those needs in some way. "Caregiving," engaging in the "hands-on" daily tasks

of care, is the aspect most frequently focused on in the literature. And their fourth

phase, "care-receiving," emphasizes that care happens within the context of a two-

way relationship rather than as a one-way dispersal of services.

Francesca Cancian and Stacey Oliker (2000, 2) also used a nurturance frame to

define care. They explained that care is "feelings of affection and responsibility

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 69

combined with actions that provide responsively for an individual's personal needs or well-being, in a face-to-face relationship." In a sense, this definition is Tronto and Fisher's (1990) phases compressed into one sentence. Both include the same essential elements-feelings, responsibility, responsive action, and relationship.

These are the qualities that have been identified by this group of scholars as the defining characteristics of care as an activity, whether that activity is happening within a family or within another institution, in a paid or an unpaid context.

One of the characteristics of sound conceptualization is its utility in empirical study. Paula England, Michelle Budig, and Nancy Folbre (2002, 459) have done some excellent work using a nurturance conceptualization of care to study the

devaluation of caring labor in the economy. They define caring labor as "work that provides a face-to-face service and develops the capabilities of the recipients." Because their goal is to apply the concept of care work to census occupational cate- gories in a study of the labor market, this formulation is necessarily somewhat nar-

rower than those in the more theoretical work discussed above. But the definition contains the same essential focus on responsive action in the context of personal relationship. That is, the theoretical starting point is the same.

The list of occupations that England and her colleagues (2002) ultimately in- cluded in their study contains many jobs that are commonly considered care work, such as nurses, teachers, and child care workers. However, it also includes physi- cians, therapists, and social workers-better paid, more highly educated workers

whose inclusion as care workers is more controversial. The authors specified a con- ceptual frame and then made decisions about what to include based on the fit with that conceptual frame rather than on commonsense understandings. This kind of conceptual specification is an important process if we are to link theoretical and empirical work in the field. The current study, which is engaged in such a linkage, builds on the work of England and her colleagues to operationalize the nurturance approach to care work.

Reproductive Labor

The nurturance framework has a strong presence in the care literature, has been relatively well theorized, and is increasingly the direction of choice for many care scholars. However, there is by no means a clear consensus among scholars and activists in the field around defining care in this way. For example, Marjorie DeVault's (1991) work on the organization of feeding care work within families includes a whole range of activities, some of which are relational, and others of which-such as shopping and doing dishes-are not. Special conferences and journal issues on care work often include the work of Pierrette Hondagneu-Sotelo (2001), Mary Romero (1992), and others who conduct research about domestic service, a job that often involves primarily cleaning, cooking, and other nonrelational tasks.

What these examples show is the presence within the care literature of another theoretical strand related to earlier feminist conceptions of domestic labor and

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

70 GENDER & SOCIETY / February 2005

social reproduction. While there is strong overlap and connection between these

fields, the conceptualization of reproductive labor is distinct from that of nur- turance in important ways. And as I will show in this article, the differences between the two frameworks have important implications for examining race and class inequalities in the distribution of care work.

The concept of reproductive labor was originally framed as a way of bringing

women's unpaid work in the home into the discourse of Marxist economics

(Hansen and Philipson 1990). The category allowed socialist-feminists to describe

women's domestic role in terms of its relationship to production, as activity that supports and maintains productive work and productive workers. In his classic arti-

cle on this topic, Wally Secombe (1974, 9) explained that when "the housewife acts directly upon wage-purchased goods and necessarily alters their form, her labour

becomes part of the congealed mass of past labour embodied in labour power." The language here is quite different from that of nurturant care. The discussion is not of

feelings or of relationship but rather of the important role of this work in the econ-

omy. Reproductive labor is defined as work that is necessary to ensure the daily

maintenance and ongoing reproduction of the labor force. The extensive literature

that emerged as feminists began to explore this topic shared as a common thread

this emphasis on making visible the critical role of domestic labor in maintaining

the productivity of current workers as well as providing the care needed to prepare

future workers for the labor force (Boydston 1990; Dalla Costa 1972; Hartmann

[1976] 1990).

While originally formulated as a way to describe unpaid work, later feminists have expanded the concept to bridge the unpaid and paid spheres. For example,

Barbara Laslett and Johanna Brenner (1989, 383) defined social reproduction as

including "various kinds of work-mental, manual, and emotional-aimed at pro-

viding the historically and socially, as well as biologically, defined care necessary

to maintain existing life and to reproduce the next generation." In their formulation,

the person doing the work is not necessarily a housewife, but the essential defini-

tional element of the role of the work in social reproduction remains.

Like the nurturance framework of care, the concept of reproductive labor has

been operationalized and used in empirical study by numerous scholars. Evelyn

Nakano Glenn used the concept in a fascinating study of the historical racial divi-

sion of reproductive labor. She started with a definition of social reproduction as

"the array of activities and relationships involved in maintaining people both on a

daily basis and intergenerationally" (1992, 115). She then identified broad census occupational categories that represent important segments of waged reproductive

labor. In the contemporary context, these occupations include food preparation and

service, health care service, cleaning and building services, and personal services.

The boundaries of reproductive labor are increasingly hard to define in a service

economy in which fewer and fewer workers are engaged in production in the classic

Marxist sense. However, Nakano Glenn did limit the service jobs included as social

reproduction to those most directly affecting either the basic daily needs of workers

or the raising and educating of the next generation. The current study builds on the

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 71

long theoretical tradition outlined above and Nakano Glenn's important empirical work to operationalize reproductive labor as an alternative framework for studying care.

Comparing the Two Frameworks

Theoretically, there is certainly some overlap between these conceptualizations of reproductive labor and of nurturant care. However, there are also important dis-

tinctions. Most significantly, the theoretical starting point is quite different. In for-

mulations of nurturant care, the emphasis is on the nature of the activity as inher-

ently relational, while conceptualizations of reproductive labor focus on the role of

the work in maintaining and reproducing the labor force. The assumption of the centrality of relationship to care, the foundation of the nurturance framework, is absent in discussions of reproductive labor. Some of the tasks of social reproduc- tion are relational, and some are not. While later theorizing about reproductive labor has become more explicit in its acknowledgment of the important emotional work involved in some of these tasks, the concept itself is much broader and less dependent on emotional connection than care viewed through a nurturance lens.

When the two concepts are applied to empirical study in the context of paid mar-

ket care, a reproductive labor definition also captures a broader universe of workers

than a nurturance framework. Some workers-such as child care providers and nurses-are included by both formulations. However, reproductive labor also includes a whole range of occupations-such as food preparation, housecleaning, and laundry-that are excluded by a nurturance frame. Figure 1 shows that in prac- tical terms, the occupations included by a nurturance definition of care make up a smaller subset of those occupations captured by a broader definition of repro- ductive labor.

In this article, I use empirical labor market data to show that these two funda- mentally different ways of conceptualizing care yield significantly different pic- tures of the gender and racial distribution of care work and thus have distinct theo- retical, empirical, and political consequences. Before describing the study in more

detail, I will briefly review the existing literature on the racial division of care work to contextualize the findings of this study more clearly.

THE RACIAL DIVISION OF CARE WORK

It is well established by economic and sociological research that the current labor market in the United States is stratified, segmented into various sectors that provide workers with grossly unequal wage levels and access to opportunities for advancement. Within this multilayer labor market, interlocking systems of gen- der and racial oppression act to concentrate women and people of color in those occupations that are lower paying and lower status (Amott and Matthaei 1996; Danzinger and Gottschalk 1995; Doeringer and Piore 1985; Hartmann [1976]

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

72 GENDER & SOCIETY / February 2005

Reproductive Labor

Nurturance

Figure 1: Universe of Workers Captured by the Reproductive Labor and Nurturance

Frameworks

1990). Evelyn Nakano Glenn (1992) and Dorothy Roberts (1997) have both written

about how the dynamics of gendered and racialized occupational segregation apply specifically to the division of reproductive labor.

Nakano Glenn (1992) has shown that both historically and in a contemporary

context, white women tend to be the "public face" of reproductive labor, concen- trated in those jobs that require the most interaction. In contrast, women of color are

disproportionately represented in the "dirty, back-room" jobs such as maids and

kitchen workers. Roberts's (1997) argument that there is a racialized division

within the category of reproductive labor between what she calls "spiritual" and

"menial" housework parallels Nakano Glenn's findings. Roberts argued that the

spiritual dimensions of domestic labor (in Marxist terms, the emotional labor) have been the province of white women, while women of color have been disproportion-

ately expected to perform the more menial (or manual) tasks. In light of the recent emphasis in the care literature on reframing care as

nurturance, defined by emotion and relationship, I found these arguments particu-

larly interesting. The work of Nakano Glenn (1992) and Roberts (1997) suggests that in fact it is where reproductive labor is seen to lack the need for emotional skills

and relational interaction that women of color are concentrated. Furthermore, these

back-room jobs are even lower paying than those more public reproductive labor

occupations in which white women are more concentrated.

As I began to unravel the implications of using a nurturance or reproductive

labor framework in an analysis of care work, the challenge posed by the work of

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 73

Nakano Glenn (1992) and Roberts (1997) became the basis of a nagging question: Were scholars inadvertently excluding the experiences of women of color and poor

women by placing increasing theoretical emphasis on nurturant care? Given the concerns expressed at the policy level about the racialized transfer of care work, this question seemed especially relevant and became the driving question of this article.

DATA AND METHOD

The research on which this article is based is part of a larger historical study of market care work over the course of the twentieth century. The current article uses

2000 census data obtained through the Integrated Public Use Microdata Series (IPUMS) at the University of Minnesota (Ruggles et al. 1997).2 The IPUMS project has provided researchers with unprecedented access to a wealth of historical and contemporary census data. The 2000 IPUMS sample used in this study is a nation- ally representative sample of the population of the United States comprising more than 2.8 million individuals.

To determine which of these individuals were considered care workers by each definition, I created two dummy variables. One variable indicates whether the

worker's occupation falls into a broad definition of reproductive labor, and the other, whether it is included in the smaller subset of jobs defined by a nurturance

framework. Using the work of Nakano Glenn (1992) as a guide,3 I determined the following criteria for inclusion of an occupation as reproductive labor:

* Work that maintains daily life (physical or mental health, food preparation and ser- vice, cleaning, personal care) or

* Work that reproduces the next generation (care of children and youth).

The criteria for inclusion as nurturance were compiled from the work of England and her colleagues (England 1992; England, Budig, and Folbre 2002):

* The job should entail face-to-face service to clients, not managers or other employees; * The face-to-face service provision should constitute a major part of the worker's time;

and

* The face-to-face service provision must develop the human capabilities of the recipi- ent-these are defined to include physical and mental health, physical skills, cogni- tive skills, and emotional skills.

During the coding process, I consulted both the census Index of Industries and Occupations (U.S. Bureau of the Census 1992) and the Dictionary of Occupational Titles (U.S. Department of Labor 1977) to get accurate information about the con- tent of each occupation in order to make an appropriate judgment relative to its inclusion in each category. My goal was to operationalize the two conceptualiza-

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

74 GENDER & SOCIETY / February 2005

tions and apply them to empirical study rather than rely on commonsense under-

standings of care work to determine inclusion or exclusion.

Table 1 provides a full list of the census occupational titles included in each cate-

gory for this study. Both columns together represent the full universe of reproduc-

tive labor workers, while the left column shows those occupations also included by

a nurturance framework. Teachers, health care workers, child care providers, and

social workers are engaged in tasks that function in the economy as reproductive

labor and are also considered to have the essential relational element of nurturant

care. However, there is another group of occupations-primarily in cleaning and

food service-that is captured by a broader definition of reproductive labor but

excluded when care is framed as nurturance. Because the definition of care is still contested within the field, some of my deci-

sions about what to include as care work will inevitably trigger disagreement.

There are some important discussions to be had around the margins of what is or is

not included as care work, and this study is intended to contribute to that dialogue.

However, it is also important to point out that the basic argument of this article

remains unchanged by the inclusion or exclusion of any particular occupation or

even group of occupations. Within the 2000 data used in this study, there are more

than 1.3 million individuals in the labor force sample. The largest occupation in the study (elementary and middle school teachers) includes about 30,000 individuals,

and the vast majority of the occupational categories are only about one-tenth that

size. Because of the relatively small size of each occupation compared to the overall

sample size, the basic findings remain unchanged by the addition or exclusion of

single occupational categories.

In addition to creating the two care variables, I also created a new race variable

by combining the existing census measure of race with another variable measuring

Hispanic background. The 2000 census contains very detailed racial/ethnic catego-

ries and also breaks down Hispanic background by country of origin. However, my

goal in this article is to test broad concepts at a national level rather than conduct a

detailed analysis of racial/ethnic stratification in particular local contexts. For that

purpose, I chose to use broad racial/ethnic characterizations and to include people of Hispanic origin as a separate category. This decision is a strategic one based on

the realities of racial/ethnic stratification in the United States rather than on the

inherent conceptual validity of the categories.

Once I had identified the care workers in the sample, I calculated composite sta-

tistics to compare the characteristics of occupations captured by each conceptual-

ization of care work when taken as a group. For example, I calculated the overall

median wage of those occupations considered reproductive labor. Then, by calcu- lating the same statistic for nurturant care occupations as well as for those repro- ductive labor occupations not considered nurturant, I was able to get a comparative perspective on wages in care work depending on how care work is defined. Similar

composite statistics were calculated to measure and compare gender and racial/

ethnic distributions. This methodological approach allowed me to test empirically

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 75

TABLE 1: Care Occupations in the 2000 Census

Nonnurturant Reproductive Labor

Psychologists Counselors Social workers

Miscellaneous community and social service specialists

Clergy

Directors, religious activities and education Religious workers, all other

Preschool and kindergarten teachers Elementary and middle school teachers Secondary school teachers

Special education teachers Other teachers and instructors

Teacher assistants Chiropractors

Dentists

Optometrists

Physicians and surgeons Physician assistants

Podiatrists

Registered nurses

Audiologists

Occupational therapists Physical therapists Radiation therapists

Recreational therapists Respiratory therapists

Speech language pathologists Therapists, all other Health diagnosing and treating practi-

tioners, all other

Dental hygienists

Emergency medical technicians and paramedics

Licensed practical nurses and licensed vocational nurses

Nursing, psychiatric, and home health aides Occupational therapist assistants and aides

Physical therapist assistants and aides

Massage therapists

Dental assistants Medical assistants and other healthcare

support occupations

Child care workers

Personal and home care aides

Recreation and fitness workers Residential advisors

Chefs and head cooks

First line supervisors/managers of food preparation workers

Cooks

Food preparation workers

Combined food preparation and service workers

Counter attendants Waiters and waitresses

Food servers, nonrestaurant

Dining room and cafeteria attendants and

bartender helpers Dishwashers

Food preparation and serving-related workers, all other

First-line supervisors/managers of house- keeping and janitorial workers

Janitors and building cleaners Maids and housekeeping cleaners Barbers

Hairdressers, hairstylists, and cosmetologists

Laundry and dry-cleaning workers

Nurturance

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

76 GENDER & SOCIETY / February 2005

TABLE 2: Care Workers in the Labor Force, 2000

Reproductive Labor

Total Nonnurturant

Labor Reproductive Reproductive

Force Labor Nurturance Labor

Number of workers 138,754,378 28,392,425 17,689,907 10,702,518

Percentage of labor force 100.0 20.5 12.7 7.7

NOTE: Estimates of total population values based on 2000 Integrated Public Use Microdata Series 1 percent sample.

the implications of the two theoretical models of care in the paid labor market using a large, nationally representative sample.

ANALYSIS AND RESULTS

Table 2 reveals the enormous practical implications of the conceptual distinction

between reproductive labor and nurturance as frameworks for care. When care is

conceptualized as reproductive labor, more than 28 million workers are considered

care workers, representing about 20 percent of the labor force. By contrast, defin-

ing care as nurturance includes slightly more than 17.5 million workers (12.7 per- cent of the labor force). The empirical difference for those studying care workers is

certainly not trivial-more than 10 million workers.

In Table 3, I have borrowed a measure from Teresa Amott and Julie Matthaei

(1996) to examine the representation of women of various racial/ethnic groups

among care workers. The relative concentration is the ratio of a group's representa- tion in a particular sector relative to that group's representation in the labor market as a whole. For example, the first column of Table 3 shows that 33.5 percent of the

workers in the labor market overall are white, non-Hispanic women. As indicated

in the next column, white, non-Hispanic women make up 49.2 percent of the repro-

ductive labor sector. The relative concentration measure of 1.5 (49.2/33.5) captures

this overrepresentation. A relative concentration of exactly 1 would indicate per-

fectly proportional representation, and a ratio of less than 1 would indicate an

underrepresentation of that group. The relative concentration of 1.5 indicates that

white, non-Hispanic women are represented among reproductive laborers at a rate

50 percent higher than their representation in the labor force overall. As seen in

Table 3, all groups of women are overrepresented in care work no matter how it is defined. However, a closer look at racial/ethnic breakdowns does reveal some

important patterns.

Given the suggestion by Nakano Glenn (1992) and Roberts (1997) that white

women tend to be concentrated in the reproductive labor jobs that require the most

public interaction and also have an emotional/spiritual dimension, one might ex-

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 77

TABLE 3: Women, by Race, as a Percentage of All Workers, 2000

Reproductive Labor

Total Nonnurturant

Labor Reproductive Reproductive

Force Labor Nurturance Labor

White, non-Hispanic womena (%) 33.5 49.2 57.4 35.6

Relative concentrationb - 1.5 1.7 1.1 Black, non-Hispanic

women (%) 5.6 9.6 10.6 7.9 Relative concentration - 1.7 1.9 1.4 Asian/Pacific Islander, non-

Hispanic women (%) 1.8 2.2 2.3 2.0 Relative concentration - 1.3 1.3 1.2

Hispanic women (%) 4.5 7.2 6.0 9.1 Relative concentration - 1.6 1.3 2.0 Other/multiracial women (%) 1.1 1.7 1.7 1.7 Relative concentration - 1.5 1.5 1.5 Total women (%) 46.5 69.9 78.0 56.4 Relative concentration - 1.5 1.7 1.2

NOTE: Dashes indicate that the relative concentration measure is not applicable in that col- umn, which represents the whole labor force. a. The number of white, non-Hispanic women as a percentage of the sector. For example, the first column indicates that 33.5 percent of the workers in the labor force are white, non-Hispanic women. Likewise, 49.2 percent of workers in the reproductive labor sector are white, non- Hispanic women, and so forth. b. A ratio of the percentage of white, non-Hispanic women in a particular sector to the percent- age of white, non-Hispanic women in the labor force. A ratio of exactly 1 means that white, non- Hispanic women are represented proportionately in this sector, less than 1 that women are underrepresented in that sector, and more than 1 that they are overrepresented (Amott and Matthaei 1996).

pect to see an especially high representation of white women among those jobs included by a nurturance framework of care, which is based on parallel defining characteristics. In fact, Table 3 does show that white women are represented at a much higher rate (1.7) among those jobs included by a nurturance conceptualiza- tion than among those more menial back-room jobs of reproductive labor excluded by such a formulation (1.1).

Although one would expect to find the opposite pattern for women of color, the

picture that emerges from the data is somewhat more complicated. For Hispanic women, the pattern is in fact the exact opposite of that of white women. The relative

concentration of this group among those reproductive labor jobs excluded by a nurturance definition of care is much higher (2.0) than among those jobs consid- ered nurturant (1.3). However, Black women are more highly overrepresented among nurturant care occupations (1.9) than among the nonnurturant jobs of repro-

ductive labor (1.4). And the pattern for Asian/Pacific women is nearly even, with

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

78 GENDER & SOCIETY / February 2005

relative concentrations of 1.3 and 1.2, respectively, among nurturant and non-

nurturant occupations.

The level of overrepresentation of Black women in nurturance jobs is a rela-

tively new phenomenon and is largely due to two distinct occupational shifts: First,

the opening up of nursing and teaching, both large occupational categories, to

Black women and, second, the growth of the category of attendants and orderlies in

hospitals and nursing homes, a category in which Black women are heavily over-

represented. It is important to note that the increasing representation of Black

women among nurturant care occupations has not been accompanied by an equally

large drop in their concentration among nonnurturant reproductive labor workers.

So, while white women are barely overrepresented in nonnurturant jobs, Black women remain concentrated there at the same time as their representation in parts

of the nurturance sector has increased. In fact, all women of color are overrepre-

sented at considerably higher rates than white women among reproductive labor

occupations that are excluded by a definition of care based on nurturance. A theo- retical focus on nurturance certainly appears to privilege the experiences of white

women over the more varied care work experiences of women of color.

Like most hierarchies in the labor market, this one is also linked to wage inequal-

ities. The second column of Table 4 shows that the median hourly wage for workers

across the reproductive labor sector4 in 2000 was $11.05, considerably lower than the median for the economy as a whole ($12.98). However, when care is defined as nurturance, the median hourly wage for care workers is actually higher ($14.40) than the overall median.5 There is a wide range of variation in wage levels within each of these groups, but when taken together, the occupations excluded by a nurturance frame are a much lower-paying set of jobs than those included.6

The third row of Table 4 provides another way of looking at the data that support

this general picture. In the labor force as a whole in 2000, 9.4 percent of workers earned less than the federal minimum wage ($5.15 per hour). By any standard, this

is a very-low-wage group of workers. When care is conceptualized broadly as reproductive labor, 14.1 percent of care workers earned less than the federal mini- mum wage, a considerably higher proportion than in the overall labor force. How-

ever, when care is limited to a nurturance frame, the proportion of minimum wage workers drops to 8.8 percent. By contrast, the group of reproductive labor workers

excluded by such a conceptualization is almost a quarter minimum wage earners.

That amounts to 2.3 million very-low-wage workers who are not included as care workers by a nurturance formulation-again, not a trivial number. Reproductive labor as a sector of the economy includes a wide range of occupations with an equally wide range of wage levels. It is important to note that a large part of the very

bottom of that wage distribution is excluded when care is defined by a narrower nurturance frame.

Looking at the list of occupations included in each category, some would argue

that these wage inequalities are not surprising. Many of the jobs included by a nurturance framework are professionalized occupations that require significantly

higher levels of formal education than the more menial jobs that make up the rest of

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 79

TABLE 4: Wages of Care Workers, 2000

Reproductive Labor

Total Nonnurturant

Labor Reproductive Reproductive

Force Labor Nurturance Labor

Median hourly wagea ($) 12.98b 11.05 14.40 7.69 Median annual earningsc ($) 32,000 26,500 32,000 18,000 Workers earning less than federal minimum waged (%) 9.4 14.1 8.8 23.2

a. For all workers who reported working during previous year. Calculated using individual as unit of analysis.

b. Dollar amounts are in contemporary (2000) dollars. c. For full-time, year-round workers (reported working more than 49 weeks last year, 35 hours a week or more).

d. For all workers who reported working during previous year. Federal minimum wage in 2000 was $5.15 per hour.

reproductive labor. That is, in fact, precisely the concern raised by this analysis. Defining care work as nurturance focuses attention on this group of more pro- fessionalized jobs at the expense of the very-low-wage workers doing the back- room work of social reproduction.

CONCLUSIONS/IMPLICATIONS

FOR THEORY AND POLICY

An empirical application of the nurturance and reproductive labor conceptual- izations of care work shows that in the context of paid care, a theoretical focus on nurturance privileges the experiences of white women and excludes large numbers

of very-low-wage workers. At the beginning of this article, I posed the question of whether scholars are inadvertently excluding the experiences of women of color and poor women by placing increasing theoretical emphasis on nurturant care. According to this analysis, the answer appears to be yes. The next question is, What are the implications of this exclusion?

First, feminists have learned from experience that the development of sound and

complete theoretical understandings of complex problems cannot be built on the experiences of one particular group of women. Exploring the devaluation of

nurturance is clearly an important path for theoretical development and empiri- cal study. However, the present study and the work of Nakano Glenn (1992) and Roberts (1997) suggest that the devaluation of nurturance is not the only problem.

In fact, when one examines racial/ethnic hierarchies within reproductive labor, jobs

that share the defining characteristics of nurturing care-relational interaction and an emotional dimension-are as a group more dominated by white women, more professionalized, and higher paying than their nonnurturant counterparts. Develop-

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

80 GENDER & SOCIETY / February 2005

ing a theoretical framework for understanding the links between care and inequali-

ties requires the integration of the study of these various hierarchies and the inter-

acting roles of gender, race, and class in shaping the distribution of care work. In the policy arena, feminists have identified as an area of major concern the

racialization of low-wage care work and the transfer of labor from white middle- and upper-middle-class women to poor women and women of color. By focusing

on framing care as nurturance, scholars may obscure one of the major axes along which the racial division of reproductive labor has historically occurred. There is no doubt that racialized occupational hierarchies exist within those care jobs cap-

tured by a nurturance definition. However, if one of the major transfers of labor has

historically been to concentrate women of color in the back-room jobs of reproduc-

tive labor, eliminating that group of jobs from consideration significantly limits

scholars' ability to fully analyze the problem of racialized transfers of care work. It

is also important for scholars to consider that framing these labor transfers as occur-

ring between groups of women assumes that the work is inherently the responsibil-

ity of women and neglects the benefits to men, other family members, and society

as a whole of care performed by paid workers.7

Finally, the group of scholars and activists that has coalesced around care has the

explicit goal of social change. There have been numerous calls for a "care move-

ment" (Stone 2000) or even for using the concept of care as the basis for a new sys-

tem of ethics (Tronto 1998). Moving in the theoretical direction of framing care as

nurturance has particular implications for the way in which parts of this movement develop. In the context of paid care work, a movement to organize care workers

based on a nurturance framework would include 10 million fewer workers than one

based on a broader definition of reproductive labor. And it is important to note that

this group includes a high concentration of women of color as well as a large num-

ber of very-low-wage workers.

In addition, one of the strategies suggested by a care movement is to revalue

nurturance by making it more visible, emphasizing the skills required by the work,

and aiming for ongoing professionalization of these occupations. In the context of a

racial hierarchy in which women of color bear much more of the burden of those

reproductive labor jobs that do not have the characteristics of nurturant care, a

movement to revalue nurturance could have the unintended effect of making their

jobs even more invisible and devalued. Framing the value of certain occupations in

terms of the emotional and relational skills required-and even professionalizing

those skills-may risk further devaluing those "menial" jobs that are not perceived to require those skills.

The potential consequences of developing a theoretical formulation of care

that is framed exclusively in terms of nurturance are counter to feminist goals of

developing an integrated analysis of race, class, and gender and addressing the

policy and equality problems engendered by the division of care work along racial/

ethnic lines. The challenge for care scholars is to think broadly about ways to

bring together current work on nurturance with other important theoretical and

empirical research-including the literature on reproductive labor and on racial

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

Duffy / REPRODUCING LABOR INEQUALITIES 81

segregation-to create a more inclusive and nuanced model of the very complex interactions between care and inequalities.

NOTES

1. This label comes from the work of Paula England (1992).

2. For more information about the sample used in this study and about the Integrated Public Use Microdata Series in general, see www.ipums.umn.edu.

3. I relied heavily on Evelyn Nakano Glenn's (1992) formulation of reproductive labor and Paula

England, Michelle Budig, and Nancy Folbre's (2002) formulation of nurturant caring labor when operationalizing these definitions, particularly in relation to census occupational categories. I thank

them all for their clarity of thought and thank Nakano Glenn and England for their willingness to give me

feedback during the coding process. The final coding decisions were mine, and by giving them credit for

their influence, I do not mean to imply responsibility for the results.

4. Median wages for each occupational group were calculated using individuals as the unit of analysis.

5. In early presentations of this research, a number of people questioned whether these findings were

due primarily to the inclusion of high-wage physicians as care workers in the nurturance category. The

four highest-wage occupations (physicians, dentists, optometrists, podiatrists) make up about 6 percent

of the nurturance category, and when they are excluded, the median for the group overall drops to $13.84,

leaving the findings basically unchanged.

6. I have included the median annual earnings of full-time, year-round workers in the table for refer-

ence, but in the discussion, I have focused on the hourly wage as the more appropriate measure because

of the large numbers of part-time workers in many care occupations.

7. Thank you to one of the anonymous reviewers for raising this very important point.

REFERENCES

Abel, Emily K., and Margaret K. Nelson. 1990. Circles of care: An introductory essay. In Circles of care:

Work and identity in women's lives, edited by Emily K. Abel and Margaret K. Nelson. Albany: State University of New York Press.

Amott, Teresa, and Julie Matthaei. 1996. Race, gender and work: A multicultural economic history of women in the United States. Boston: South End.

Boydston, Jeanne. 1990. Home and work: Housework, wages, and the ideology of labor in the early republic. New York: Oxford University Press.

Cancian, Francesca M., and Stacey J. Oliker. 2000. Caring and gender. Thousand Oaks, CA: Pine Forge Press.

Dalla Costa, Mariarosa. 1972. The power of women and the subversion of the community. Bristol, UK: Falling Wall Press.

Danzinger, Sheldon, and Peter Gottschalk. 1995. America unequal. New York: Russell Sage. DeVault, Marjorie L. 1991. Feeding the family: The social organization of caring as gendered work. Chi-

cago: University of Chicago Press.

Doeringer, Peter, and Michael Piore. 1985. Internal labor markets and manpower analysis. New York: M. E. Sharpe.

England, Paula. 1992. Comparable worth: Theories and evidence. New York: Aldine De Gruyter. England, Paula, Michelle Budig, and Nancy Folbre. 2002. Wages of virtue: The relative pay of care

work. Social Problems 49:455-73.

Folbre, Nancy. 2001. The invisible heart: Economics and family values. New York: New Press.

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms

82 GENDER & SOCIETY / February 2005

Hansen, Karen V., and Ilene J. Philipson. 1990. Women, class, and the feminist imagination: A socialist- feminist reader. Philadelphia: Temple University Press.

Harrington, Mona. 1999. Care and equality: Inventing a new family politics. New York: Alfred Knopf.

Hartmann, Heidi I. [1976] 1990. Capitalism, patriarchy, and job segregation by sex. In Women, class, and

the feminist imagination: A socialist-feminist reader, edited by Karen V. Hansen and Ilene J. Philipson. Philadelphia: Temple University Press.

Heymann, Jody. 2002. Can working families ever win? Helping parents succeed at work and caregiving. Boston Review, March.

Hondagneu-Sotelo, Pierrette. 2000. The international division of caring and cleaning work. In Care work: Gender, labor and the welfare state, edited by Madonna H. Meyer. New York: Routledge.

- . 2001. Domdstica: Immigrant workers cleaning and caring in the shadows of affluence. Berke- ley: University of California Press.

Laslett, Barbara, and Johanna Brenner. 1989. Gender and social reproduction: Historical perspectives. Annual Review of Sociology 15:38 1-404.

Milkman, Ruth. 1998. The macrosociology of paid domestic labor. Work and Occupations 25 (4): 483- 510.

Nakano Glenn, Evelyn. 1992. From servitude to service work: Historical continuities in the racial divi-

sion of paid reproductive labor. Signs: Journal of Women in Culture and Society 18:1-43.

Noddings, Nel, Suzanne Gordon, and Patricia Benner. 1996. Caregiving: Readings in knowledge, prac- tice, ethics, and politics. Philadelphia: University of Pennsylvania Press.

Roberts, Dorothy. 1997. Spiritual and menial housework. Yale Journal of Law and Feminism 9:51-80.

Romero, Mary. 1992. Maid in the U.S.A. New York: Routledge.

Ruggles, Steven, Matthew Sobek, et al. 1997. Integrated Public Use Microdata Series: Version 2.0.

Minneapolis: Historical Census Projects, University of Minnesota. Available from http://www.

ipums.org.

Russell Hochschild, Arlie. 1998. Ideals of care: Traditional, postmodern, cold-modern, and warm-

modern. In Families in the U.S.: Kinship and domestic politics, edited by Karen V. Hansen and Anita Ilta Garey. Philadelphia: Temple University Press.

Secombe, Wally. 1974. The housewife and her labour under capitalism. New Left Review 83:3-24. Stone, Deborah. 2000. Why we need a care movement. The Nation, 13 March.

Tronto, Joan. 1993. Moral boundaries: A political argument for an ethic of care. New York: Routledge.

---- . 1998. An ethic of care. Generations 22:15-20.

---- . 2002. The "nanny" question in feminism. Hypatia 17:34-51.

Tronto, Joan, and Berenice Fisher. 1990. Towards a feminist theory of caring. In Circles of care: Work

and identity in women's lives, edited by Emily K. Abel and Margaret K. NJelson. Albany: State Uni-

versity of New York Press.

U.S. Bureau of the Census. 1992. Classified index of industries and occupations, 1990 Census ofpopula-

tion and housing. Washington, DC: Government Printing Office.

U.S. Department of Labor. 1977. Dictionary of occupational titles. 4th ed. Washington, DC: Govern- ment Printing Office.

Mignon Duffy is a doctoral candidate for a joint degree in Sociology and Social Policy at

Brandeis University and the Heller School. Her research interests include paid and unpaid care,

families, and education policy. In January, Mignon will join the faculty of the Sociology Depart-

ment at the University of Massachusetts, Lowell.

This content downloaded from 154.59.124.94 on Thu, 11 Apr 2019 16:01:16 UTC All use subject to https://about.jstor.org/terms