Corporations will be equally hiring men and women based on their skills and knowledge in the future.

M TRAVIS 12/16/2009 2:55:14 PM

385

THE FUTURE OF WORK-FAMILY POLICY: IS “CHOICE” THE RIGHT CHOICE? REVIEW OF WOMEN AND EMPLOYMENT:

CHANGING LIVES AND NEW CHALLENGES*

BY MICHELLE A. TRAVIS**

I. INTRODUCTION ...............................................................................385 II. WOMEN AND EMPLOYMENT: CHANGING LIVES AND NEW

CHALLENGES ..................................................................................390 III. THE “CHOICE” EQUALITY MODEL..................................................396

A. Evidence of Constrained Choice .............................................403 B. Legal and Policy Implications .................................................414

1. The Less-Is-More Choice Paradox and Other Challenges ...................................................................418

2. The Role of Reflexive Law in Regulating Choice.........425 IV. CONCLUSION ...................................................................................429

I. INTRODUCTION

“‘There’s no such thing as work-life balance[]’. . . . ‘There are work- life choices, and you make them, and they have consequences.’”

—Jack Welch, Former CEO of General Electric Co.1 At a time when a majority of American workers report experiencing

conflict in balancing their work, personal life, and family life,2 it is not

* WOMEN AND EMPLOYMENT: CHANGING LIVES AND NEW CHALLENGES (Jacqueline Scott et al. eds., 2008). ** Professor, University of San Francisco School of Law. B.A., Cornell University; J.D., Stanford Law School. 1. Naomi Schaefer Riley, Work and Life – and Blogging the Balance, WALL ST. J., July 17, 2009, at W11; Cari Tuna & Joann S. Lublin, Welch: ‘No Such Thing as Work-Life Balance,’ WSJ.COM, WALL ST. J.: MANAGEMENT (July 14, 2009, 12:58 p.m. EDT), <http://online.wsj.com/article/SB1247 26415198325373.html> (subscription required). 2. See JERRY A. JACOBS & KATHLEEN GERSON, THE TIME DIVIDE: WORK, FAMILY, AND GENDER INEQUALITY 84-85 (2004) (describing the results of a national 1997 study finding that 55.5 percent of women and 59.8 percent of men reported experiencing “at least some conflict in balancing work, personal life, and family”).

M TRAVIS 12/16/2009 2:55:14 PM

386 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

surprising that legal scholarship on “work-family conflict” has exploded over the past decade. In the quest for legal and policy solutions to achieve better work-family balance, much of this literature has framed the debate as an issue of gender equality. Yet despite this large body of critically important work, defining the “end state” of what an appropriate or desirable work-family balance should look like remains elusive.3 More fundamentally, work-family scholars remain conflicted (or simply undecided) about what exactly is meant by “gender equality” in the context of care work and labor force participation.

Professor Michael Selmi recently highlighted this ambivalence as a significant roadblock for continued progress in the work-family policy arena.4 In summarizing the work-family debate, Selmi identified two broad perspectives that have dominated discourse on how to define the “gender equality” objective in the work-family context.5 The first perspective focuses on achieving equal outcomes for women and men in the paid labor market.6 Under this view, benchmarks of achievement would include reducing occupational segregation, integrating women into the top of the workplace hierarchy, and eliminating the gender pay gap. Policy proposals flowing from this perspective typically focus on methods for allowing women to increase their labor force attachment, such as “equal parenting, longer school days, and greater public support for child care.”7 To the extent that equal parenting may not be readily achievable in the short-term, a related strand of the workplace equality approach focuses additionally on minimizing the labor market costs of temporary leaves for childbirth and periods of reduced working hours when caregiving demands are particularly acute, through “more and better part-time jobs, often with proportionate benefits,” and other flexible arrangements for working hours and location.8

3. See Michael Selmi, The Work-Family Conflict: An Essay on Employers, Men and Responsibility, 4 U. ST. THOMAS L.J. 573, 575 (2007). 4. See id. at 573-77. 5. See id. at 573-76. In this Review, I have modified Professor Selmi’s taxonomy modestly. Selmi actually describes three distinct perspectives, see id. at 573-74, while I combine the first two of his perspectives into a single perspective with two related strands. Either approach usefully divides the workplace-focused equality norms from the alternative approach (Selmi’s third perspective), which focuses instead on addressing society’s devaluation of work in the home. See id. I combined Selmi’s first two workplace-focused perspectives in part because that taxonomy maps more closely onto the taxonomy that is used in the text that is the subject of this book review. See infra Part III. 6. See Selmi, supra note 3, at 573-74. 7. See id. at 573. 8. See id. at 574; see also Vicki Schultz & Allison Hoffman, The Need for a Reduced Workweek in the United States, in PRECARIOUS WORK, WOMEN, AND THE NEW ECONOMY: THE CHALLENGE TO LEGAL NORMS 131, 133-34 (Judith Fudge & Rosemary Owen eds., 2006) (describing this approach in U.S. feminist legal literature).

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 387

The second broad perspective embraces “a preferred role for women as caregivers,” and focuses on achieving equal valuation of unwaged domestic care work with work performed in the paid labor market.9 Policy proposals flowing from this perspective typically focus on various types of subsidies to provide economic self-sufficiency to those performing a disproportionate amount of care work, as well as expanding maternity leave provisions, increasing the quality of low-hour part-time positions, or taking other steps to allow women to spend more time out of the labor market and engaged in care work.

Because normative and often very personal considerations inevitably influence one’s commitment to these alternative perspectives,10 some work-family scholars have attempted to bridge the divide and forge a way forward by shifting away from calls for particular substantive outcomes towards a more process-based conception of equality. This new perspective focuses on equality of choice amongst a range of combinations of market work and care work.11 This new direction in defining gender equality has a consensus-building potential within the feminist legal community because it avoids the difficult normative questions about “appropriate” gender roles and about the primary locus of a fulfilling life. On the other hand, Professor Selmi notes that this perspective remains under-theorized, sometimes retaining its ties to gender equality rhetoric by arguing that women should have access to the same choices as men, and sometimes seeming to jettison equality altogether and devolving into an unworkable assertion “that women ought to be able to do whatever they want without penalty.”12 Specifically, Selmi worries that “a desire to support all choices for women” will end up “stalemat[ing]” the work-family debate altogether.13

Another challenge for the equal choice perspective is that work-family scholars currently disagree as to how much of the existing gendered division of paid and unpaid labor already reflects freely-chosen gendered preferences. While mainstream legal feminists generally do not embrace either of the extreme ends of the choice continuum—i.e., that the gendered division of labor reflects either wholly unconstrained preferences or that it is the result of literal coercion—there is enormous room for disagreement about where reality falls within the space between those ends. Thus, one sub-debate that has long troubled work-family scholars is to what extent the

9. See Selmi, supra note 3, at 574-75; see also Schultz & Hoffman, supra note 8, at 133-34 (describing this approach in U.S. feminist legal literature). 10. See Selmi, supra note 3, at 575. 11. See id. at 575-76. 12. See id. 13. See id. at 576.

M TRAVIS 12/16/2009 2:55:14 PM

388 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

gendered division of both paid and unpaid labor represents constrained choice, and more specifically, the extent to which women’s work-family choices are more constrained (or at least differently constrained) than men’s.14 Resolution of this sub-debate has significant implications for translating an equal choice perspective into concrete policy proposals. However, unlike the normative considerations affecting one’s perspective on work-family policy, which Professor Selmi notes “are not easily influenced by data,”15 this sub-debate is, to at least some degree, an empirical question.

The centrality of this and other empirical inquiries in advancing a cohesive work-family agenda highlights the importance of the new, interdisciplinary book, Women and Employment: Changing Lives and New Challenges, edited by Jacqueline Scott, Shirley Dex, and Heather Joshi, and published by Edward Elgar Publishing Limited.16 Women and Employment is a compilation of thirteen empirical studies and policy proposals by twenty-eight authors, only one of whom is a law professor. The authors include experts in sociology, demography, statistics, economics, human geography, social policy studies, and industrial relations.17 The articles in Women and Employment were produced as part of the twenty-fifth anniversary of the United Kingdom’s Women and Employment Survey (WES), which was one of the most ground-breaking governmental surveys of women and employment ever undertaken at the time.18 The WES surveyed a nationally representative group of nearly six thousand women in the U.K. and, using retrospective work histories and longitudinal data, provided a comprehensive view of women’s economic activity within that country.19 In Women and Employment, contemporary researchers analyze the results of a wide array of new data to determine how much progress has been made towards achieving women’s workplace equality in the twenty-five years since the WES, and to suggest future directions for legal and policy reform.

Although the empirical work and policy proposals in Women and Employment are focused on the U.K., the book has much to offer American work-family advocates and scholars. The book’s impressive use of a

14. See infra Part III.A. 15. See Selmi, supra note 3, at 575. 16. WOMEN AND EMPLOYMENT, supra note *. 17. See id. at vii-xiv. 18. See Jacqueline Scott et al., Introduction: Changing Lives and New Challenges, in WOMEN AND EMPLOYMENT, supra note *, at 1, 1-2; see also Jean Martin & Ceridwen Roberts, Putting Women on the Research Agenda: The 1980 Women and Employment Survey, in WOMEN AND EMPLOYMENT, supra note *, at 109, 113-29 (describing the WES). 19. See Scott et al., supra note 18, at 2; Martin & Roberts, supra note 18, at 113-15.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 389

variety of research methodologies is a valuable model for empiricists seeking to converge upon a better understanding of women’s relationship to work and care. For work-family scholars, the nuanced analysis of data along not only gender lines, but also with respect to class, race, ethnicity, cultural background, age, and family structure is an important reminder that our work must remain responsive to the heterogeneity of women workers. Some of the book’s studies explicitly incorporate comparative data from both the United States and the European Union to help assess women’s relative workplace success.20 And, to the extent that the U.K.’s demographic, workplace, and regulatory environments are similar to ours, even the non-comparative data may provide useful insights – or at least working hypotheses – about various factual assumptions underlying American work-family scholarship.

This Review will highlight one such potential contribution by analyzing what the book’s research might add to our understanding of constrained choice in women’s and men’s work-family balance solutions. Specifically, this Review will identify some of the book’s empirical findings that may help us make a more accurate causal attribution for the gendered division of labor along the choice continuum between the extremes of preference and coercion. More broadly, it will examine what the book’s findings and policy analysis might reveal about the viability of an equal choice model as a path forward in the American work-family “stalemate.” These are by no means the book’s only contributions; rather, they are used to illustrate the book’s significant potential to inform contemporary American work-family debates.21

Part II of this Review briefly summarizes Women and Employment and highlights its major strengths and weaknesses as a resource for work- family scholars. Part III uses the emerging equal choice model as a frame for evaluating some of the book’s empirical work and policy analysis. Part III.A synthesizes some of the book’s major research findings regarding constraints on work-family decisionmaking to help American work-family scholars achieve greater consensus on the underlying question of the extent to which gendered preferences may explain the existing gendered division of labor. Because this new research reveals a significant disconnect between actual and ideal work-family balance solutions, and it

20. See, e.g., Jane Lewis, Work-Family Balance Policies: Issues and Development in the UK 1997-2005 in Comparative Perspective, in WOMEN AND EMPLOYMENT, supra note *, at 268; Jacqueline Scott, Changing Gender Role Attitudes, in WOMEN AND EMPLOYMENT, supra note *, at 156. 21. I will leave to social scientists the task of evaluating the methodological soundness of the studies. My focus is on the studies’ potential contribution to the legal feminism literature on work- family conflict.

M TRAVIS 12/16/2009 2:55:14 PM

390 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

demonstrates that work-family “choices” are constrained in meaningfully different ways along both gender and class lines, these findings validate consideration of what a “choice model” might offer future work-family policy debates. Part III.B therefore discusses the insights that the book’s policy articles provide about whether and how a choice model might be translated into legal and policy initiatives. This Part considers the general question of whether a choice-based agenda can meaningfully retain its ties to gender equality theory, and it highlights some of the more specific challenges that such an approach might face. It ends by suggesting one potential direction for future research to build upon the valuable work in Women and Employment by considering an equal choice model within the emerging field of reflexive legal regulation.

II. WOMEN AND EMPLOYMENT: CHANGING LIVES AND NEW CHALLENGES

The U.K.’s Women in Employment Survey (WES), published in 1984, has been described as “path-breaking,” “‘radical,’” and “‘revolutionary,’” and it forms the backdrop for the work in Women and Employment.22 For those unfamiliar with the WES, chapter four is a good place to begin the book, as it describes the context, objectives, and methodology of this comprehensive governmental study.23 The rather immodest goals of the WES included not only understanding “the place of employment in women’s lives,” but also revealing how women’s employment “related to the wider issue of men’s and women’s roles in the family.”24

One important methodological contribution of the WES was to go beyond cross-sectional surveys to collect retrospective work histories.25 These histories provided longitudinal data of women’s patterns in and out of the labor market as their domestic responsibilities changed.26 As a result of this “lifetime perspective,” the WES debunked two widely-held beliefs at the time: that most women leave the labor market permanently after having children, and that labor market participation was unimportant to most women’s lives.27 By documenting that the great majority of women return to employment after having children (and were doing so increasingly earlier in their children’s lives), and by showing that market work was highly important to most women (both for personal satisfaction and

22. See Martin & Roberts, supra note 18, at 109, 127. 23. See generally id. 24. Id. at 112. 25. Id. 26. Id. at 113-16. 27. Id. at 115, 118, 124, 128.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 391

because most families were highly dependent upon women’s earnings), the WES permanently secured a place for “women as workers” on future research and policy agendas.28 Women and Employment was the result of grant funding to support additional qualitative and quantitative studies at the local, national, and cross-national levels to understand how women’s employment has changed in the twenty-five years since the WES, and to “look ahead to likely changes to come.”29

The articles in Women and Employment are grouped into four Parts. The first two Parts attempt to uncover ways in which women’s work and family lives have changed. Part I, titled “Women and Employment: Assessing Progress on Equality,” includes three empirical studies that evaluate how women currently are faring in the U.K.’s labor market. The first study analyzes the effects of women’s increased participation in higher education by gathering longitudinal employment data on women who obtained their first degrees in 1995.30 The second study analyzes the occupational mobility of women in different age cohorts after returning to the labor market from having a first child.31 The third study compares the labor market activity for women of different ethnicities at various stages in their lives.32 One study in Part III of the book should be added to this initial set, as it documents women’s increasing levels of labor market participation after having a child, primarily in part-time positions.33

These labor market studies are complemented by two additional works in Part II, titled “Dynamics of Employment and Family Across the Life Course,” which focuses on related changes outside the workplace,34 and by one similar study in Part IV. One of these studies documents the rise in cohabitation, delayed marriage, and childbirth outside of marriage.35 Another highlights the enormous increase in the transnational movement of workers and the complicated effects of diaspora on women’s labor market

28. Id. 116-18, 124, 128-29 (emphasis added). 29. See Scott et al., supra note 18, at 2; see also WOMEN AND EMPLOYMENT, supra note *, at xv (describing the grant funding). 30. Kate Purcell & Peter Elias, Achieving Equality in the Knowledge Economy, in WOMEN AND EMPLOYMENT, supra note *, at 19. 31. Shirley Dex et al., Changes in Women’s Occupations and Occupational Mobility Over 25 Years, in WOMEN AND EMPLOYMENT, supra note *, at 54. 32. Angela Dale et al., Ethnic Differences in Women’s Labour Market Activity, in WOMEN AND EMPLOYMENT, supra note *, at 81. 33. Susan McRae, Working Full-Time After Motherhood, in WOMEN AND EMPLOYMENT, supra note *, at 179. 34. Part II also includes an article that describes the WES. See Martin & Roberts, supra note 18. 35. John Ermisch, The New Dynamics of Family Formation and the Explosion of Childbearing Outside Marriage, in WOMEN AND EMPLOYMENT, supra note *, at 133.

M TRAVIS 12/16/2009 2:55:14 PM

392 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

mily life.

attachment, occupational mobility, and caregiving responsibilities.36 The final study shifts from social to psychological phenomena, by measuring changes in gender role attitudes between 1988 and 2002, a time in which women’s labor force participation rates in the U.K. rose to rates comparable to the U.S. (although with higher rates of part-time work in the U.K.).37

After providing a nuanced picture of women’s contemporary work and family lives, Part III of the book, aptly titled “Work-Life Balance,” then seeks to understand how women have combined the two and to measure the “conflict” that women are experiencing in attempting to balance the competing demands of care work and labor force participation. Two studies assess how class affects the availability of workplace flexibility, decisions regarding the number of hours engaged in paid labor, and self-reports of work-family conflict levels.38 A third study measures how the household division of labor affects women’s allocation of paid and unpaid work, and is affected by the male earner’s relative number of work hours and compensation.39

The final article in Part III is the first to take a policy turn. That article reviews the U.K.’s regulatory interventions into the work-family arena since the late 1990’s, compares those interventions to approaches in continental European countries and the U.S., and hypothesizes about the relationship between various policy initiatives and women’s labor force attachment.40 Part IV completes the book with three forward-looking policy articles that recommend “Ways Forward” in regulating the relationship between work and fa 41

Although the articles in Women and Employment are focused on the U.K., much of the book’s analysis uses comparative U.S. data, which provides a useful metric for assessing American women’s work-family

36. Linda McDowell et al., Migration, Employment and Gender Divisions of Labour, in WOMEN AND EMPLOYMENT, supra note *, at 329. 37. Scott, supra note 20, at 156. 38. Colette Fagan et al., Class Differences in Mothers’ Work Schedules and Assessments of Their ‘Work-Life Balance’ in Dual-Earner Couples in Britain, in WOMEN AND EMPLOYMENT, supra note *, at 199; Rosemary Crompton & Clare Lyonette, Mothers’ Employment, Work-Life Conflict, Careers and Class, in WOMEN AND EMPLOYMENT, supra note *, at 213. 39. Susan Harkness, The Household Division of Labour: Changes in Families’ Allocation of Paid and Unpaid Work, in WOMEN AND EMPLOYMENT, supra note *, at 234. 40. Lewis, supra note 20, at 268. 41. Jill Rubery, Women and Work in the UK: The Need for a Modernisation of Labour Market Institutions, in WOMEN AND EMPLOYMENT, supra note *, at 289; Simon Deakin & Colm McLaughlin, The Regulation of Women’s Pay: From Individual Rights to Reflexive Law?, in WOMEN AND EMPLOYMENT, supra note *, at 313; Susan Himmelweit, Policy on Care: A Help or a Hindrance to Gender Equality?, in WOMEN AND EMPLOYMENT, supra note *, at 347.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 393

experiences. One of the most revealing examples is a study on changes in gender role attitudes. This cross-national study compares survey responses from men and women in 1988, 1994, and 2002, to a series of questions about gender role ideology.42 Answers that supported non-traditional gender roles were characterized as “egalitarian.”43 Somewhat surprising was the finding that a lower percentage of U.S. respondents held egalitarian gender role attitudes than in many other countries.44 Even more surprising was evidence of a significant and uniquely-American retreat from egalitarian views during this period.45

For example, one question asked respondents if they thought that a husband’s job was to earn income and a wife’s job was to take care of children.46 The percentage of U.S. respondents that provided an egalitarian answer (i.e., “no”) rose from 51.5 percent in 1988 to 59.4 percent in 1994, and then dropped back down to 52.5 percent in 2002.47 The percentage of egalitarian responses in 2002 was much higher in other countries surveyed, including the U.K. (64.5 percent), West Germany (61.1 percent), Ireland (69.3 percent), the Netherlands (67.5 percent), and Spain (66.4 percent), and no other country evidenced a drop in support for gender egalitarian roles since 1994.48 Interestingly, during the same period in which U.S. respondents increased their endorsement of traditional gender roles, they also modestly increased their endorsement of the view that both the husband and the wife should contribute to household income (57.6 percent in 1994 to 58.3 percent in 2002).49 It is thus not surprising that Arlie Russell Hochschild’s phrase, “the second shift,” which describes women’s combination of paid work with disproportionate care work, continues to resonate with so many American women.50

In addition to providing a rich source of comparative data, one of the greatest strengths of Women and Employment is that many of the book’s empirical inquiries are consciously informed by an analysis of class, race, ethnicity, cultural background, age, and family structure, which much of American work-family scholarship has been criticized for ignoring.51 In

42. Scott, supra note 20, at 162-73. 43. Id. at 163. 44. See id. at 169, tbl. 6.2. 45. See id. at 167-69, tbl. 6.2. 46. Id. at 169, tbl. 6.2. 47. Id. 48. Id. 49. Id. 50. See generally ARLIE RUSSELL HOCHSCHILD, THE SECOND SHIFT: WORKING FAMILIES AND THE REVOLUTION AT HOME (1989) (originally coining the phrase, “the second shift”). 51. See, e.g., Nancy E. Dowd, Bringing the Margin to the Center: Comprehensive Strategies for

M TRAVIS 12/16/2009 2:55:14 PM

394 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

Professor Laura Kessler’s recent “classcrit” essay, she particularly laments the continued lack of interest that American legal feminist literature shows towards class analysis.52 One of Kessler’s concrete proposals to help work- family scholars incorporate class into their work is to make greater use of sociology, which “has a long tradition of attention to social class, not as an afterthought but as a main concern.”53 As Kessler notes, sociological methods “are highly effective at revealing subtle but significant differences within groups,”54 which can reveal the complex ways that various institutional and social structures influence women’s behavior and, in turn, can orient legal scholars towards the most appropriate mix of legal reforms. The empirical studies in Women and Employment uphold this tradition of bringing social class issues to the forefront. For this reason, the book is an excellent resource for work-family scholars who would like to take Kessler’s suggestion seriously as a way to better understand the relationship between structural discrimination and gender-based economic inequality.

While each of the individual articles in Women & Employment thus has much to offer American work-family scholars, readers may be disappointed that neither the book’s editors nor the individual authors do more to make the book greater than the sum of its parts. The book includes a concise Introduction that provides context for the empirical studies, compares core data on women’s labor force participation in the U.K. today and at the time of the WES, and describes the book’s organization.55 Beyond that, neither the editors nor the individual authors engage each others’ work or respond to any specific, shared research questions. There is no attempt to synthesize the findings in the various empirical studies in each Part of the book, nor do the authors of the policy articles directly incorporate or rely on the empirical findings in the other articles to ground their recommendations.

Additionally, although the book uses gender equality as a loose frame for measuring the progress that has taken place in the twenty-five years

Work/Family Policies, 73 U. CIN. L. REV. 433, 434 (2004) (urging work-family scholars to pay closer attention to racial and class inequalities, and to marginalized caregivers); Laura T. Kessler, Getting Class, 56 BUFF. L. REV. 915, 915-19 (2008) (arguing that “much legal feminist literature remains uninterested in class analysis,” and providing examples of work-family scholarship that pays insufficient attention to class); Michael Selmi & Naomi Cahn, Women in the Workplace: Which Women, Which Agenda?, 13 DUKE J. GENDER L. & POL’Y 7, 7-8 (2006) (arguing that a disproportionate amount of work-family literature “has focused on a small segment of women – typically professional women, lawyers, professors or corporate executives,” and has ignored issues of class). 52. See Kessler, supra note 51, at 915. 53. Id. at 929-32. 54. Id. at 930. 55. Scott et al., supra note 18.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 395

since the WES, the authors never define the “equality” metric against which they are evaluating the new empirical findings. As Professor Selmi’s description of very different perspectives on “gender equality” in the work- family context demonstrates, effectively evaluating progress and using data to inform policy initiatives depends upon the lens through which one peers. While the researchers in the original WES study explicitly adopted the objective of finding ways to reduce female unemployment, it is unclear whether that or other objectives were informing the various follow-up studies in Women and Employment.

Another recent book published in the U.K. provides one useful model for a more integrated approach to empirical and policy studies. In 2004, Catherine Hakim of the London School of Economics authored the second edition of Key Issues in Women’s Work: Female Diversity and the Polarisation of Women’s Employment, which similarly synthesizes an impressive array of data on women’s participation in the U.K.’s market economy, with similar comparative elements.56 As in Women and Employment, the core chapters of Hakim’s book provide detailed empirical findings about women’s participation in various types of waged and unwaged labor, the characteristics of the part-time workforce, the extent of sex segregation in jobs and labor market segmentation, the gender pay gap, the labor market effects of childbirth and motherhood, and attitudes towards the sexual division of labor and work-life balance.57 What Hakim’s book adds that is absent from Women and Employment, however, is a theoretical framework to help the reader evaluate the data and upon which to ground the book’s policy recommendations.

Hakim begins her book by identifying four perspectives that have dominated European thought on the causes of women’s subordinate social and economic position.58 These include Steven Goldberg’s theory that psycho-physiological processes render male dominance and patriarchy inevitable,59 Heidi Hartmann’s theory of men’s collective organization to further their own interests relative to women,60 Gary Becker’s rational exchange theory to explain the sexual division of labor as a form of

56. CATHERINE HAKIM, KEY ISSUES IN WOMEN’S WORK: FEMALE DIVERSITY AND THE POLARISATION OF WOMEN’S EMPLOYMENT (2d ed. 2004) [hereinafter HAKIM, KEY ISSUES]; see also generally EMPLOYMENT POLICY AND THE REGULATION OF PART-TIME WORK IN THE EUROPEAN UNION: A COMPARATIVE ANALYSIS (Silvana Sciarra et al. eds., 2004) (using data on part-time work in the European Union as a tool for examining the role that European Community rules and procedures play in the development of national social policies within the member states). 57. See HAKIM, KEY ISSUES, supra note 56, chs. 2-6. 58. See id. at 3-17. 59. Id. at 4-8. 60. Id. at 8-11.

M TRAVIS 12/16/2009 2:55:14 PM

396 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

allocative efficiency,61 and Hakim’s own theory of gendered work and lifestyle preferences within modern liberal societies.62 In each section of her book, Hakim assesses how well each of these four theories can account for various empirical findings.63 One risk of this approach, of course, stems from the well-documented information-processing bias of focusing disproportionately on evidence that confirms one’s own theory and of discounting inconsistent or disconfirming evidence.64 Not surprisingly, Hakim’s ultimate conclusion is that her own “preference theory” best explains the diverse set of research results that are summarized in her book.65 Nevertheless, the theoretical frames provide a useful, concrete metric with which to consider and assess the book’s complex empirical data.

The authors and editors of Women and Employment were perhaps even better-situated to provide such a theoretical frame for their book because, unlike Hakim, they were not so obviously entrenched as long-term advocates of any particular definition of equality or theory of causation for explaining women’s second-class economic citizenship. The remaining sections of this Review will offer an initial glimpse into how one model of gender equality and its corresponding theory of causation could be used to cull additional value from the already inherently rich empirical and policy analysis that Women and Employment offers. Specifically, Part III of this Review explores what the book’s findings reveal about a “choice” perspective, which appears to be implicitly lodged between its lines.

III. THE “CHOICE” EQUALITY MODEL

As Professor Selmi suggests, moving the American work-family debate forward requires a more precise definition of “gender equality,” if gender equality is to remain a driving force behind policy reform.66 As noted above, Selmi has identified three different perspectives on what a

61. Id. at 11-13. 62. Id. at 13-17. Catherine Hakim has articulated her “preference theory” more fully in earlier works. See CATHERINE HAKIM, WORK-LIFESTYLE CHOICES IN THE 21ST CENTURY: PREFERENCE THEORY (2000); Catherine Hakim, Lifestyle Preferences as Determinants of Women’s Differentiated Labor Market Careers, 29 WORK & OCCUPATIONS 428 (2002); Catherine Hakim, Public Morality Versus Personal Choice: The Failure of Social Attitude Surveys, 54 BRIT. J. SOC. 339 (2003). 63. See HAKIM, KEY ISSUES, supra note 56, at 18. 64. See generally Raymond S. Nickerson, Confirmation Bias: A Ubiquitous Phenomenon in Many Guises, 2 REV. GEN. PSYCHOL. 175 (1998) (describing how people’s information processing is affected by a confirmatory bias). For a list of social science work on the confirmation bias, see Confirmation Bias/Confirmatory Bias, <http://confirmation-bias.behaviouralfinance.net/> (last visited Oct. 29, 2009). 65. See HAKIM, KEY ISSUES, supra note 56, at 214. 66. See Selmi, supra note 3, at 575.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 397

gender-equality objective might mean in the work-family context: (1) equal outcomes between women and men in the labor market (which would best be achieved by an equal division of domestic work); (2) equal valuation of unwaged care work with labor market work; and (3) the newly-emerging equal choice perspective.67 Although Women and Employment does not explicitly identify any of these perspectives as its metric when analyzing women’s progress over the past twenty-five years, one of the most interesting articles at least highlights this dilemma when evaluating the success of existing government interventions and considering various policy reforms.

The author of this particular article, Jane Lewis, a Professor of Social Policy at the London School of Economics and Political Science, shares Professor Selmi’s view that a cohesive policy strategy requires tackling the “vexed issue of what might constitute gender equality in terms of work- family balance.”68 Lewis identifies three potential definitions, which mirror Selmi’s taxonomy.69 She characterizes the first two perspectives – equal labor market results, and equal valuation of care work – as outcomes- focused definitions.70 The former definition, in its strongest form, would seek to use government intervention to achieve the outcome of “an equal division of paid and unpaid work between men and women: a citizen worker/carer model.”71 The latter definition, in its strongest form, would seek instead to use government intervention to achieve the outcome of “reward[ing] women’s disproportionate amount of care work.”72 Lewis notes that one risk of the perspective favoring equal labor market results is that policy-makers may construe it “in terms of making women more like men, especially in respect of the male career model.”73 On the other hand, one risk of the equal value perspective is “perpetuating [care] work as women’s responsibility.”74 Both of these perspectives, she notes, also share the additional challenge of being inevitably “value-laden” in the work- family context, in which “the whole idea of balancing work and family may be a euphemism for competing ideologies about childrearing.”75

67. See supra notes 5-13 and accompanying text. 68. See Lewis, supra note 20, at 274 (noting that “[i]t is very difficult to define gender equality in the context of work-family balance policies”). 69. See id. at 274-77. 70. See id. at 274. Lewis links these two perspectives to the “age-old” debate over “equality-as- sameness or equality-as-difference” that periodically has divided feminist legal thought. Id. 71. Id. 72. Id. 73. Id. 74. Id. 75. Id. at 274-75.

M TRAVIS 12/16/2009 2:55:14 PM

398 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

Lewis frames the third perspective – the choice approach – as process- focused rather than outcomes-oriented.76 By implicitly locating the choice approach within the burgeoning literature on process-based interventions into the labor market,77 Lewis appears less skeptical than Selmi about the potential of a choice approach to move the work-family debate forward.78 Lewis and Selmi also appear to disagree about the role that gender equality would play in a choice perspective on work-family policy. Selmi seems to suggest that a choice model that seeks more than just ensuring women access to the same options that men currently possess for balancing paid and unpaid work will ultimately stall work-family initiatives.79 Lewis, in contrast, envisions a consensus-building policy objective aimed at maximizing choices for both women and men with respect to work-family balance decisions.80 This explains Lewis’s decision to label this perspective a “genuine choice” rather than an “equal choice” policy aim.81

Lewis acknowledges significant challenges in translating a genuine choice perspective into concrete governmental policies, some of which are discussed in Part B of this Review below. More generally, she recognizes that a choice model necessarily presumes an understanding of the current gendered division of paid and unpaid labor as a reflection of constraints rather than preferences. If women’s current relationship with the labor market is simply “a matter of revealed preferences,” notes Lewis, “then there are no policy issues in respect of gender equality.”82 Thus, any form of a choice model requires assessing the extent to which existing work- family balance decisions reflect constrained choices, the sources of those constraints, and whether women are experiencing more (or at least

76. See id. at 274-77. 77. See infra Part III.B.2. 78. See Lewis, supra note 20, at 275 (suggesting that a genuine choice approach may be “strategic for those concerned about gender equality”). 79. See Selmi, supra note 3, at 576 (noting that “[f]rom the equality perspective, the primary question is whether women are afforded the same, or equivalent, choices as men,” and that “a desire to support all choices for women” has “stalemated” work-family debates). Selmi agrees with Lewis that “we can, and should work to change or expand” the choices available to both men and women, but as an overarching policy objective, he argues that “we should not treat all choices as equally valid or valuable.” Id. More specifically, Selmi parts company with Lewis in concluding that “[i]f we are concerned about women’s equality . . . , we should not focus on policies that allow more women to stay at home, or spend more time out of the workplace.” Id. Cf. Lewis, supra note 20, at 275 (arguing that “[g]iven the difficulties in interpreting what people want by way of work-family balance, together with the difficulties in prioritising any form of gender equality within most governments’ work-family balance policies, it may be strategic for those concerned about gender equality, as well democratic, to focus on inputs and to aim for policies that maximise genuine choice for men and women in respect of work-family balance” (emphasis in original)). 80. See Lewis, supra note 20, at 275-76. 81. See id. 82. Id. at 273.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 399

different) constraints than men. These questions have been a longstanding source of unease, if not

outright disagreement, for both American and British work-family scholars.83 In the American work-family literature, some feminist scholarship is built upon the assumption that women’s economic disadvantage arises in significant part from their deep commitment to caregiving.84 Whether that commitment is the result of inherently gendered differences in nurturing skills and interest, or the product of gender-role socialization, these scholars often argue that work-family objectives should focus on valuing women’s caregiving, rather than on seeking complete gender integration in paid and unpaid work.85 In contrast, other feminist scholars fear the reification of gender-based work patterns and instead highlight the perceived role played by employers’ practices, institutional norms, and other external constraints in creating and entrenching the

83. See JOAN WILLIAMS, UNBENDING GENDER: WHY FAMILY AND WORK CONFLICT AND WHAT TO DO ABOUT IT 38 (2000) (describing the debate within American work-family scholarship and identifying scholars associated most strongly with the extreme positions of preference (i.e., “agency”) and coercion (i.e., “the exclusion of agency”)); see also Michelle A. Travis, Equality in the Virtual Workplace, 24 BERKELEY J. EMP. & LAB. L. 283, 286-87 (2003) (describing the ongoing debate regarding the extent of constrained choice in work-family decisions). For information on the parallel debate in the U.K., see infra notes 93-97 and accompanying text. 84. See Vicki Schultz, Life’s Work, 100 COLUM. L. REV. 1881, 1893, 1900 (2000) (explaining that “some feminist scholars argue that women’s economic disadvantage arises from their primary commitment to their families – rather than from sexist dynamics in labor markets and firms,” which assumes that “women’s domestic orientation is fixed by the time we enter the labor force”); Travis, supra note 83, at 286 (describing this strand of work-family scholarship); see, e.g., Joan Williams, “It’s Snowing Down South”: How to Help Mothers and Avoid Recycling the Sameness/Difference Debate, 102 COLUM. L. REV. 812, 819, 824-26 (2002) (stating that “a significant proportion of women’s economic disadvantage is attributable to their family roles,” and using work-family literature, family law literature, and gender theory to demonstrate the persistence of women’s “commitments to caregiving and other traditionally feminine gender performances”). 85. See Schultz, supra note 84, at 1884 (describing this feminist position as premised on the view “that it is women’s position within families, rather than the workworld, that is the primary cause of women’s economic disadvantage, and hence should be the primary locus for redistributive efforts”); Travis, supra note 83, at 286 (describing the emphasis of these scholars on “not devalu[ing] women’s caregiving,” nor prioritizing complete gender integration in waged and unwaged work); Williams, supra note 84, at 828-29 (describing “care work advocates” who “reject[] the strategy of empowering women by requiring them to abandon traditionally feminine gender performances”); Joan Williams, From Difference to Dominance to Domesticity: Care as Work, Gender as Tradition, 76 CHI.-KENT L. REV. 1441, 1456 (2001) (listing authors who have proposed various programs for providing subsidies for care work); see, e.g., Martha Ertman, Love and Work: A Response to Vicki Schultz’s Life’s Work, 102 COLUM. L. REV. 848, 851-52 (2002) (arguing for work-family proposals that treat care work as “equally important” to paid work); Martha Albertson Fineman, Contract and Care, 76 CHI.-KENT L. REV. 1403, 1406 (2001) (arguing that because “caretaking work supplies an important and essential public good,” more public support should be provided for the needs of caretakers); Laura T. Kessler, The Attachment Gap: Employment Discrimination Law, Women’s Cultural Caregiving, and the Limits of Economic and Liberal Legal Theory, 34 U. MICH. J.L. REV. 371, 372 (2001) (arguing for a greater “focus on the fundamental importance and value to society and to women of women’s caregiving labor”).

M TRAVIS 12/16/2009 2:55:14 PM

400 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

connection between women and care work.86 By attributing much of the gendered nature of care work to sexist labor market dynamics, social controls within the workplace, and other behavioral constraints, these latter scholars urge a greater focus on achieving truly integrated work in both the paid labor market and at home.87

Not surprisingly, scholars who situate themselves closer to the preference end of the choice continuum are more likely to align themselves primarily with an equal valuation perspective, while scholars who situate themselves closer to the coercion end of the choice continuum are more likely to align themselves primarily with an equal labor market perspective.88 This means that for a new “choice perspective” to successfully build greater consensus within the work-family community, more needs to be done to resolve the underlying “preference versus constraint” debate.

This preference versus constraint debate has garnered a disproportionate amount of attention in the context of work-family decisions made by highly-educated, professional women, some of whom have left the labor market to engage in full-time childrearing. The question of the extent to which these women’s decisions reflect wholly unconstrained preferences took on an iconic role in work-family debates with Lisa Belkin’s 2003 New York Times article, “The Opt-Out Revolution,” which painted a rosy picture of professional women’s willing

86. See, e.g., Schultz, supra note 84, at 1883-84, 1892-1907 (arguing that women’s economic disadvantage results in significant part “from sexist dynamics in labor markets and firms,” and concluding that workplace disadvantages cause the gendered nature of carework rather than the other way around); Vicki Schultz, Telling Stories About Women and Work: Judicial Interpretations of Sex Segregation in the Workplace in Title VII Cases Raising the Lack of Interest Argument, 103 HARV. L. REV. 1749, 1816, 1824, 1841 (1990) (arguing that “employers create women’s job preferences” in the ways that they structure higher-paying jobs); Travis, supra note 83, at 286-88 (describing this strand of work-family scholarship and using sociological data to reveal how employers are structuring telecommuting arrangements to entrench gender inequalities in both paid and unpaid work); JOAN C. WILLIAMS ET AL., CTR. FOR WORKLIFE LAW, UNIV. OF CAL. HASTINGS COLLEGE OF LAW, “OPT OUT” OR PUSHED OUT?: HOW THE PRESS COVERS WORK/FAMILY CONFLICT: THE UNTOLD STORY OF WHY WOMEN LEAVE THE WORKFORCE 7 (2006), available at <http://www.worklifelaw.org/pubs/OptOut PushedOut.pdf> (last visited Aug. 1, 2009) (arguing that “most mothers do not opt out; they are pushed out by workplace inflexibility, the lack of family supports, and workplace bias against mothers”). 87. See Travis, supra note 83, at 287 (describing this perspective); see, e.g., Schultz, supra note 84, at 1883-1907 (describing the importance of increasing women’s labor market attachment as a means for achieving greater gender equality); Selmi, supra note 3, at 597 (emphasizing the importance of policies that create greater workforce attachment for women, rather than policies that enable “women to spend more time out of the workplace”). 88. See Schultz, supra note 84, at 1884, 1899-1907, 1928 (noting that one’s view of the primary cause of women’s economic disadvantage as either “women’s position within families” or women’s position within “the workworld” affects one’s view on the “primary locus for redistributive efforts,” because if women’s sociobiological commitment to childcare is the primary cause then solutions should focus on increasing the value of unpaid work, but if workplace structures are the primary cause then solutions should focus on challenging sex bias in the labor market).

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 401

exodus from the workplace to take on the preferred role of full-time mother.89 Belkin’s article has generated fierce critique by those who perceive greater constraints on these women’s decisions,90 and it has generated an entire new lexicon for debates over the choice continuum (e.g., “opt out” versus “pushed out,” “off-ramping” and “on-ramping,” etc.). The fervor of this debate likely will be rekindled by the more recent public remarks of Jack Welch, the former CEO of General Electric Co., in which he attached himself to the preference end of the continuum when discussing professional women’s attempts to climb the corporate ladder.91 “‘There’s no such thing as work-life balance,’” said Welch, “‘There are work-life choices, and you make them, and they have consequences.’”92 This narrow but highly visible slice of the choice debate highlights the deeper divides over the role of gendered preferences and gendered constraints that has long existed within American work-family literature.

In the U.K., a similar debate has been taking place, fueled in part by Professor Catherine Hakim, whose recent book is described above.93 Hakim has staked out a position closest to the preference end of the choice continuum by arguing that women’s secondary position in the labor market is primarily a function of their personal preferences, with structural constraints becoming increasingly less important within modern liberal societies.94 Specifically, Hakim argues that women are freely able to choose their preferred number of working hours in countries with flexible and deregulated labor markets like the U.S. and the U.K., where she

89. Lisa Belkin, The Opt-Out Revolution, N.Y. TIMES, Oct. 26, 2003, § 6 (Magazine), at 44; see also Jane Gross, Forget the Career. My Parents Need Me at Home, N.Y. TIMES, Nov. 24, 2005, at A1 (suggesting that women who leave the workforce to engage in full-time caregiving “may see leaving a high-powered career as an opportunity, not a sacrifice”); Laura T. Kessler, Keeping Discrimination Theory Front and Center in the Discourse over Work and Family Conflict, 34 PEPP. L. REV. 313, 320 (2007) (noting that Belkin’s article generated “a great deal of recent attention” by promoting “[t]he idea that women are autonomous, unsituated actors fully responsible for their secondary position in the workforce”). 90. See, e.g., WILLIAMS ET AL., supra note 86, at 4-6; Rachel J. Littman, Building a Sustainable Model for the Legal Industry, 29 PACE L. REV. 317, 324-26 (2009); Pamela Stone, Panel One: Professional Women and Work/Life Conflict, 27 WOMEN’S RTS. L. REP. 27, 27-31 (2006) [hereinafter Stone, Panel One]; Alison I. Stein, Note, Women Lawyers Blog for Workplace Equality: Blogging as a Feminist Legal Method, 20 YALE J.L. & FEMINISM 357, 366-68 (2009). See generally PAMELA STONE, OPTING OUT? WHY WOMEN REALLY QUIT CAREERS AND HEAD HOME (2007) [hereinafter STONE, OPTING OUT] (analyzing interviews of professional women who left the workforce after having children and assessing the constrained nature of women’s decisions to exit the labor force). 91. Tuna & Lublin, supra note 1. 92. Id. Notably, Welch does not recognize that the “consequences” may be precisely what render women’s “choices” highly constrained, nor does he recognize the role that workplace practices, policies, and norms have in narrowing the options from which women may “choose.” 93. See supra notes 56-65 and accompanying text. 94. See HAKIM, KEY ISSUES, supra note 56, at 13-17; see also Lewis, supra note 20, at 273 (describing Hakim’s position).

M TRAVIS 12/16/2009 2:55:14 PM

402 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

contends women have secured equal employment opportunities.95 While Hakim acknowledges that “institutions, laws, customs, national policies and cultural constraints continue to shape and structure behaviour,” she asserts that, at least in “prosperous modern societies,” these factors are secondary to the primary determinant of women’s individual (and heterogeneous) preferences.96 Under Hakim’s view, in which “genuine choices are now open to . . . the vast majority of women,” there would be no role for a choice approach to gender equality in the work-family policy arena. Hakim’s work on her self-described “preference theory” has garnered extensive critique by European researchers whose causal attribution theories regarding external constraints on work-family balance decisions situate them much closer to the other end of the choice continuum.97

In the Introduction to Women and Employment, the editors acknowledge these ongoing debates, and they suggest that one objective of the book’s empirical studies is to take on the weighty task of “disentangl[ing] the influence of preferences and constraints on the different priorities of men and women concerning family and employment.”98 The authors of the individual empirical articles, however, deal with this fundamental causal attribution question largely as an afterthought rather than as a central theme in their analysis. Thus, while the book’s data has the potential to provide significant insights on the question of constrained choice, there is no part of the book that attempts to synthesize the data around this issue. The following Parts of this Review will take a first step in that direction.99

95. See Crompton & Lyonette, supra note 38, at 217 (describing Hakim’s thesis); see also HAKIM, KEY ISSUES, supra note 56, at 15. 96. HAKIM, KEY ISSUES, supra note 56, at 15. Hakim argues that “once genuine choices are open to them, women choose between three different lifestyles: adaptive, work-centered or home-centered.” Id. at 14-15. Adaptive women “prefer to combine employment and family work within their lives without giving a fixed priority to either”; work-centered women “are focused on competitive activities in the public sphere”; and home-centered women “prefer to give priority to home and family life.” Id. 97. See Lewis, supra note 20, at 273 (describing the response to Hakim’s thesis); see, e.g., Crompton & Lyonette, supra note 38, at 217 (challenging Hakim’s thesis by arguing that women’s work-family choices are “made in the context of the opportunities available,” and that “this context remains substantially stratified by both class and gender”). 98. Scott et al., supra note 18, at 2. 99. While this Part of the Review explores what the book’s empirical work might contribute to understanding the need for or implementation of a choice model for work-family reform, it is not clear which, if any, of the authors of the individual articles in Women and Employment endorse an “equal choice” or “genuine choice” perspective. While some of the articles indicate support for unconstrained work-family choices as a metric of equality, others seem to focus solely on assessing equality of outcomes in the labor market. In this part of my Review, I consider the choice model simply to provide an evaluative lens through which to view the book’s data, and thereby to demonstrate one way in which the sum of the book could be greater than its constituent parts. Other work-family scholars similarly could use the book’s data to test alternative equality models or causal attribution hypotheses.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 403

Part A highlights a few of the book’s empirical findings that support a notion of constrained choice, rather than unconstrained preference, to explain the existing gendered division of paid and unpaid labor. This evidence of a disconnect between actual and preferred work-family combinations, along with some corroborating evidence in American work- family literature, at least legitimates consideration of a choice model of equality as a potential work-family policy aim. Part B therefore returns to the general question of whether and how a choice model could be translated into meaningful policy objectives, and it considers some of the major stumbling blocks for expanding the available options for combining care work with labor market participation. Part B then addresses one specific form of legal intervention – what the authors in Women and Employment refer to as “reflexive law” – and it considers what role that particular approach might play in a choice model of work-family reform.

A. Evidence of Constrained Choice

One of the strengths of Women and Employment is the varied methodological approaches and research inquiries used in the book’s set of empirical studies. In combination, this data helps demonstrate a variety of ways in which women’s work-family balance decisions are shaped and constrained by workplace practices, role expectations, socioeconomic status, and education.100 This evidence may help work-family scholars converge upon a narrower band of space along the choice continuum (between the endpoints of preference and coercion) in their assumptions about the primary causes of the gendered division of labor. This Part will highlight some of the evidence that the book provides.

The first type of evidence relies upon women’s self-reports of their desired balance between paid labor and care work. By comparing these self-reports to women’s actual conduct, several of the studies reveal a disconnect between actual and ideal work-family allocations, suggesting that work-family decisions are not a straightforward reflection of gendered preferences. In one study, researchers interviewed a group of college- educated women and men at three different points in their first decade of work after graduation in the U.K.101 Before having children, the women and men had remarkably similar work orientations and career

100. In addition to the articles described in this Part of the Review that highlight these types of constraints, one interesting article also studied the ways in which cultural background and family structure affects women’s work-family balance decisions. See Dale et al., supra note 32. 101. Purcell & Elias, supra note 30, at 22.

M TRAVIS 12/16/2009 2:55:14 PM

404 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

ibility.

aspirations.102 For the women, their “skills, knowledge and career development” made up “important dimensions of their identities,” and neither the men nor the women regarded parenthood “as an alternative career for women.”103 Nevertheless, the women ended up accommodating their careers to a greater extent than the men when they later became parents.104 Rather than indicating a change in preferences upon the arrival of children, the women’s explanations for this greater accommodation revealed instead “[t]he significance of options available and the context within which choices were made.”105

Specifically, the researchers found a variety of ways in which “gendered attitudes and expectations” of employers and others affected the women’s work-family decisions.106 Even before they became parents, many of the women earned less than their male partners, which affects the decision about which partner will make greater career accommodations for childrearing within dual-career households. The researchers found a gender wage gap and sex-segregated occupations beginning immediately upon graduation and increasing over time, in part due to employers’ discriminatory conduct, including recruiting men and women with “androgynous occupational skills” into “‘gender appropriate’” job vacancies.107 Women also reported instances of discriminatory treatment by employers when they became pregnant, as well as more general resistance to workplace flex 108

For some women, these pressures were compounded by dynamics within their own partnerships, including navigating their partners’ demanding careers in typically high-level jobs.109 When asked whose career was more important, both men and women most commonly answered that both partners’ careers were equally important, but an interesting difference emerged in couples with disparate earners.110 While women generally accorded priority to their male partners’ careers if the male partner was the higher earner, the men generally would not prioritize their female partners’ careers, even if the female partner earned

102. Id. at 34. 103. Id. at 45-46. 104. See id. at 38-42. 105. Id. at 40 (describing the “reflexiveness of work-life decisions”). 106. Id. at 33. 107. Id. at 20-32. 108. See id. at 38-42. 109. See id. at 44-45; see also Crompton & Lyonette, supra note 38, at 219 (finding that “[n]ot only do full-time managerial and professional women work, on average, longer hours than women in the other occupational classes, but they are also more likely to be in partnerships with similar men”). 110. See Purcell & Elias, supra note 30, at 34-36.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 405

ore than twice as likely to remain in full-time than were women who reported that their husbands or

a ese views either in full or in part. Id.

lso infra notes 124-35 and

pra note 86, at 30 (describing the study).

omen lawyers,” and observing that “[d]escribing family responsibilities

significantly more.111 At a minimum, these results demonstrate that highly educated women’s work-family decisions “reflect a complex balance of preference and constraint,” with many women’s choices marked by deep “ambivalence.”112 On the whole, the researchers suggest that the evidence is more consistent with women “accommodating to practical obstacles,” than with women having “a different orientation to employment, career development, [or to] the centrality of family roles.”113

These findings are consistent with reports provided by many high occupational status American women who similarly describe their work- life decisions as accommodating the realistic obstacles of workplaces built around the norm of “ideal workers” who can provide uninterrupted and unlimited working hours free from significant caregiving responsibilities.114 As in the U.K., highly educated women in the U.S. are more likely to be in the labor force than lower educated women, are more likely to be working long full-time hours, and are more likely to desire shorter hours.115 In a recent survey of highly educated women in business, 89 percent reported that they wanted access to reduced-hour jobs.116 Among the subset of highly-trained, professional women who do leave the workforce to engage in caregiving, another survey found that only 11.6 percent “exhibited a stable preference to be stay at home moms.”117 In contrast, 86 percent of the women cited “workplace pushes,” such as inflexible jobs, as the primary reason for their decision to leave the workforce.118 Moreover, 93 percent of surveyed women who had left their jobs in business, law, medicine, or academia wanted to return to the labor

111. See id. In another study reported in the book, researchers asked women if their husbands or partners thought that a woman’s “main job is to look after the family,” or if it was only ok for a woman to work “if it fits with family life.” See McRae, supra note 33, at 185. Women who reported that their husbands or partners would endorse neither view were m continuous employment after having children p rtners endorsed th 112. Purcell & Elias, supra note 30, at 45. 113. Id. at 33. 114. See WILLIAMS, supra note 83, at 13-39; WILLIAMS ET AL., supra note 86, at 8; Stone, Panel One, supra note 90, at 28-31. 115. See WILLIAMS ET AL., supra note 86, at 2, 13, 20, 22-23; see a accompanying text (describing the parallel findings in the U.K.). 116. See WILLIAMS ET AL., su 117. See id. at 10 (describing the study). Additionally, only 16 percent had planned all along to quit work when they had children. Id. 118. See id. at 2, 10-11 (describing the study); see also Stein, supra note 90, at 367 (describing a study in which law school graduates cited “‘personal and family responsibilities’ as the ‘single largest barrier’ to the advancement of w as a ‘barrier’ strongly suggests that women are not ‘opting out,’ but instead are being pushed out by inflexible workplace policies”).

M TRAVIS 12/16/2009 2:55:14 PM

406 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

husb

d dissa

119. See SYLVIA ANN HEWLETT, OFF-RAMPS AND ON-RAMPS: KEEPING TALENTED WOMEN ON THE ROAD TO SUCCESS 29 (2007).

force.119 As in the U.K., many highly-qualified women in the U.S. also found the challenges of their inflexible workplaces compounded by the challenges of having husbands in demanding professional jobs, given that 30 to 40 percent of highly-qualified men work fifty or more hours per week.120 In one study, two-thirds of highly-qualified women reported that their decisions to reduce their paid work hours were influenced by their

ands’ careers.121 While these studies focused on a narrow set of privileged women,

two other studies in Women and Employment provide evidence of a mismatch between women’s actual and ideal working hours across both class and educational lines within the U.K. In the first of these studies, researchers interviewed mothers of preschool or primary school children in dual-earner, urban households regarding their work schedules, workplace flexibility, and desired working hours.122 In the second of these studies, researchers synthesized a wide range of national employment and survey data, including the British Social Attitudes Series, an annual survey of up to 3600 adults conducted by the National Centre for Social Research.123 Both studies found that highly educated women, particularly those in managerial and professional jobs, were more likely to remain in the workplace full-time after having children than were lower educated women.124 Women at both ends of the class spectrum, however, reported feeling “rushed or short of time in the effort to co-ordinate the demands of their jobs and the changing needs of their children.”125 More significantly, many women in both high and low socioeconomic class categories reporte

tisfaction with their working hours, but often in opposite directions. Most of the full-time managerial and professional women reported a

120. See WILLIAMS ET AL., supra note 86, at 30, 33; see also supra notes 109-13 and accompanying text (describing the parallel findings in the U.K.). 121. See WILLIAMS ET AL., supra note 86, at 33 (describing the study). 122. Fagan et al., supra note 38, at 201. 123. Crompton & Lyonette, supra note 38, at 213. For more information on the British Social Attitudes Series, see British Social Attitudes Survey, <http://www.esds.ac.uk/government/bsa/> (last visited Aug. 1, 2009), and British Social Attitudes Survey List of Databases, <http://www.data- archive.ac.uk/findingData/bsaTitles.asp> (last visited Aug. 1, 2009). 124. See Crompton & Lyonette, supra note 38, at 213-14; Fagan et al., supra note 38, at 202-07. 125. Fagan et al., supra note 38, at 207. The second study found high levels of self-reported work- life conflict across gender and class lines as well, but it also found that work-life conflict was “significantly higher (for both men and women) for those in professional and managerial occupations compared with the other class categories,” with a non-statistically significant higher level for women in that category. Crompton & Lyonette, supra note 38, at 219. This was due both to the fact that “full-time managerial and professional women work, on average, longer hours than women in the other occupational classes,” and because “they are also more likely to be in partnerships with similar men.” Id.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 407

e result of workplace constraints rather than economic needs or de

rained in doing so by financial reasons rather than work

ofessional and managerial jobs “receive better work-life entitlements from

ould be heavily penalised,” for

time workers would be happy to work fewer hours even if it meant lf of all full-time working women with children

the capacity to choose a lower level of

desire “to reduce their hours or gain more flexibility in how they organised their hours.”126 Managerial and professional women frequently described workplace norms as constraining their ability to work reduced hours, even though their jobs provided more formal opportunities for workplace flexibility.127 These women were “acutely aware” of the need “to be visibly present and respond flexibly to the demands of their job in order to demonstrate their adherence to the ‘long hours’ organisational culture in which they were embedded.”128 Evidence of an increasing willingness of full-time workers to forgo pay in exchange for reduced work hours provides further evidence that these women’s longer-than-desired number of hours is th

sires.129 The women with lower qualifications who were in lower occupational

class categories fell more clearly into two distinct groups. Some of them, like the managerial and professional women, were working very long hours, but were doing so out of economic need, often because their male partners were earning modest incomes as well.130 Like the professional and managerial women, this subset of low-income women reported a high desire to reduce their levels of employment, but they more frequently reported being const

place norms.131 On the other hand, a much larger percentage of women in the lower

occupational class categories than in the higher class categories had either reduced their working hours to a much greater degree or had left paid

126. Fagan et al., supra note 38, at 208. 127. See id. at 202-05; see also Crompton & Lyonette, supra note 38, at 218, 220 (explaining that even though women in pr their employers,” these women explicitly recognize part-time work “as having a very negative impact on career development”). 128. Fagan et al., supra note 38, at 204. “These mothers explained that long hours were a requirement of their job and that if they negotiated shorter hours they w example, through “demotion from team leader” or “loss of bonuses.” Id. at 202. As one participant explained, negotiating for shorter hours would be “‘career suicide.’” Id. 129. See Harkness, supra note 39, at 234-35 (reporting British survey results suggesting that “around 40 per cent of all full- foregoing pay, and this figure rises to almost one-ha aged under 5”). 130. See Fagan et al., supra note 38, at 200, 203. 131. See id. at 200 (“Manual employees who work long hours are the most likely to say they do so for financial reasons, while managers and professionals usually emphasise their job commitment . . . .”); see also Crompton & Lyonette, supra note 38, at 228-29 (noting that “women who earn good salaries (and who are likely to be in partnerships with similar men) have work intensity without necessarily facing economic hardship,” while “women in less well remunerated jobs are more likely to be working because of economic need”).

M TRAVIS 12/16/2009 2:55:14 PM

408 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

desire to re-enter the work

132. See Fagan et al., supra note 38, at 200, 203, 207 (finding that part-time work “is

rs are much more likely to be in a pational categories”).

ra note 38, at 206-07.

s. See Fagan et al., supra note

make and the opportunities they have available . . .”).

employment altogether.132 These women, particularly those in manual jobs, frequently reported feeling “forced to leave their job” because of even greater workplace rigidity and labor market constraints, including that “they did not qualify for maternity leave,” “their line manager had ‘made it difficult’ for them to stay,” “their employer would not let them adjust their hours,” “they did not have access to suitable childcare,” or “they had been unsuccessful in their search for an alternative job with shorter hours or closer to home.”133 For the women at lower occupational levels who were able to secure part-time work, the arrangements were commonly for twenty hours or less per week, while the much smaller group of professional and managerial women who worked part-time typically performed twenty-five to thirty-four hours per week.134 Accordingly, while these low-income women also reported a mismatch between their desired and actual work- family balance, many more of these women reported a

place or secure additional paid working hours.135 Thus, the researchers in Women and Employment express deep

skepticism about “widespread assertions to the effect that contemporary societies are increasingly characterised by more freedom and ‘choice,’” given the “very real structures of constraint, by both class and gender.”136 Specifically, class and gender shape employment options, including work hours and schedules, which in turn shape the actions that women take “to

disproportionately concentrated in low-paid service and manual jobs, often with very short hours,” that “part-time working remains rare for women who enter management or male-dominated professions,” that “[a]mong part-timers, longer hours were worked in managerial or professional positions than at lower occupational levels,” and that “the lower-qualified mothers were more likely to have dropped out of employment” entirely as a result of workplace constraints); Crompton & Lyonette, supra note 38, at 213-14 (finding that “low- and mid-skilled mothers are more likely to reduce their employment than mothers with higher skills,” while “professional and managerial women . . . are . . . significantly more likely to work full-time,” and that “professional and managerial mothe p id employment than mothers in the lower occu 133. Fagan et al., sup 134. See id. at 203. 135. See id. at 203, 206-09. Both studies did identify a small set of women who embraced a traditional maternal view and desired to leave the labor force to engage in caregiving full-time. See id. at 208-09; see also Crompton & Lyonette, supra note 38, at 226-28. As the researchers noted, these self-reported views may be the result of realistic appraisals of their workplace constraints – for highly- qualified women, the recognition of the severe career consequences in reducing hours, and for lower- qualified women, the recognition of few career opportunities and the less rewarding nature of their jobs – rather than non-contextualized preferences. See Crompton & Lyonette, supra note 38, at 222-29. This particular option for work-family balance obviously was constrained by class, with lower-income women being less able to stay home without suffering great material los 38, at 208-09; see also Crompton & Lyonette, supra note 38, at 228-29. 136. Crompton & Lyonette, supra note 38, at 216; see also id. at 229 (“Despite claims as to the increasing significance of individualism and choice, both class and gender persist as major structural constraints shaping the decisions men and women

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 409

longer hours than desired, while more women in lo

tion gap” was highest amon

supra note 38, at 217.

ercent among women and 23 percent among women in manual jobs. Id.

percent of part-time women reported that promotions were very or fairly important.

percent

gs, a ‘lack of

improve their work-life balance.”137 While acknowledging that “men and women do indeed make choices in relation to both mothers’ employment and family responsibilities more generally,” the researchers emphasize that these choices “are always made in the context of the opportunities available to individuals and families,” and that “this context remains substantially stratified by both class and gender.”138 Unfortunately, the class differences in women’s constrained choices regarding work-life balance are likely to perpetuate and deepen class inequalities, as more women in higher occupational categories work

wer occupational categories reduce their hours significantly or leave the workforce altogether.139

This deepening class divide likely is facilitated through a class- differentiated effect on career aspirations. One of the studies in Women and Employment measured promotion aspiration levels by asking U.K. respondents how important it was to them to “move up the job ladder.”140 Among full-time workers, women in higher occupational classes expressed higher levels of promotion aspiration than women in lower occupational classes.141 In addition, within all occupational class levels, women working full-time expressed higher levels of promotion aspiration than women working part-time.142 This full-time/part-time “aspira

g professional and managerial women,143 and it nearly disappeared among women in the lowest occupational classes.144

While it is possible that the different occupational class outcomes are the result of women’s differing preferences, the researchers conclude that causation is likely working in the opposite direction. The overall lower promotion aspiration levels and greater emphasis on family obligations among women engaged in routine and manual work is likely the result of “the characteristics of working class jobs,” which provide less rewards, less advancement opportunities, and less flexibility.145 For women in the higher

137. See Fagan et al., supra note 38, at 210. 138. Crompton & Lyonette, 139. See id. at 215-16, 227. 140. See id. at 223, tbl. 9.7. 141. Id. The percentage of full-time women reporting that future promotions were very or fairly important was 55 percent among women in professional or managerial jobs, 32 p in “[i]ntermediate” job classes, 142. Id. 143. Id. at 222-23, tbl. 9.7. Among professional and managerial workers, 55 percent of full-time women and only 31 Id. at 223, tbl. 9.7. 144. Id. at 222-23, tbl. 9.7. Among manual workers, 23 percent of full-time women and 22 of part-time women reported that promotions were very or fairly important. Id. at 223, tbl. 9.7. 145. Id. at 222-27 (concluding that for “women in the lower occupational groupin

M TRAVIS 12/16/2009 2:55:14 PM

410 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

ly along class lines

ambition’ may be interpreted as reflecting an adaptation to the opportunities available”).

occupational classes, the lower career aspirations of part-time workers likely reflect their knowledge of the severe career consequences of part- time work and a “realistic adaptation” to their reduced advancement opportunities.146 For women, much more than men, high promotion aspirations are associated with increased levels of work-life conflict, which means that women indeed “face an extremely difficult ‘choice.’”147 While the researchers acknowledge that it is impossible to determine the causal direction with certainty – i.e., to determine whether “less ambitious women ‘choose’ to work part-time, or whether a lack of interest in promotion is . . . a realistic adaptation to the reduced opportunities available to part-time women” – the latter appears more consistent with women’s descriptions of feeling “torn” about deciding to move to part-time work.148 In other words, the greater work-life conflict created by higher career aspirations is often resolved by lowering one’s aspirations. Because the labor market constraints impact promotion aspirations particularly negative

, and promotion aspirations affect future employment behavior, this cycle likely perpetuates and further entrenches class divides.

While these studies in Women and Employment are thus important in revealing the constrained nature of work-family decisions, they are also important in revealing the structural supports that reduce work-family conflict. In one study, the researchers discovered a subset of women employed in the U.K. across the full range of occupational levels who described their work-life balance as “‘OK’, ‘about right’, ‘as good as it can be’ or ‘reasonably optimal.’”149 At all occupational levels, these women were primarily those working relatively short full-time hours (or relatively long part-time hours) without long commutes.150 Another study found that nearly all of the highly-qualified women who were balancing employment

146. Id. at 222, 230; cf. WILLIAMS, supra note 83, at 38 (“A subjective sense of authenticity and repose about one’s ‘choices’ may reflect no more than a decision to bring one’s life into alignment with the expectations and institutions of domesticity.”). 147. See Crompton & Lyonette, supra note 38, at 225, 230. 148. See id. at 222. In addition, the study found that “the mere fact of having a child in the household does not, in itself, affect promotion aspirations,” which are determined more significantly by sex, class, age, and employment patterns. Id. at 224. 149. See Fagan et al., supra note 38, at 209. 150. See id. Some of these women were working what was described as “part-time hours,” but the researchers found that “part-time employment was not a simple panacea for reconciling the demands of jobs and family life” because of the significant consequences in career advancement that many such positions entail. See id. Similar results were found in another study in the book, in which the lowest levels of work-life conflict were reported in households with the male working full-time and the female working part-time, but, “particularly among professional and managerial women,” the part-time arrangements came with “a very negative impact on career development.” Crompton & Lyonette, supra note 38, at 220.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 411

er week. As described above, various constraints frequ

more extreme options of subs

as in the U.K., low-hour part-time jobs are held predominantly by women,

g evidence suggesting that this preference exists for he fifteen major European Union member states).

melweit, supra note 41, at 349.

conducted in

ite-collar workers have more job flexibility and control over their work schedules than blue- ing text (describing the parallel findings in the

and mothering most successfully had been able to negotiate flexibility within the jobs they had held before they had children, often through reduced hours or a four-day week.151 Overall, it appears that the work- family balance option that is most commonly desired across both class and education lines is combining caregiving with approximately thirty to forty hours of paid work p 152

ently push women to extremes below and above this amount, in a non- class-neutral manner.

This bifurcation of working hours not only disparately affects women of different classes, but it also disparately affects women and men, as women disproportionately end up working very short hours, while very long-hour positions are disproportionately held by men.153 One study described in Women and Employment found that in dual-earner couples with two full-time workers in the U.K., women averaged over forty hours of paid work per week, with only a very marginal reduction in average hours for women with children.154 Women with children were largely unable to obtain small reductions in working hours within a full-time position, and instead had to “choose” between the

tantially shorter part-time work schedules or leaving the workforce altogether.155

Researchers have found similar results in the U.S., where the bifurcation of working hours due to inflexible workplace practices has both class effects among women and gendered effects between women and men.156 As in the U.K., low-income women in the U.S. face even more rigid workplaces than high-income women.157 While high-income women more typically feel constrained by workplace norms that demand very long hours, low-income women often face severe job consequences or job loss for even minor schedule changes due to childrearing needs.158 In the U.S.,

151. See Purcell & Elias, supra note 30, at 42-43. 152. See Lewis, supra note 20, at 273 (citin men and women with young children in all of t 153. See Him 154. See Harkness, supra note 39, at 245. 155. See id. 156. See WILLIAMS ET AL., supra note 86, at 29. Although much of this research was the 1990s and early 2000s, the more recent data in Women and Employment finding similar trends in the U.K. provides some basis for believing that the trends have continued here as well. 157. See id. at 36; see also JACOBS & GERSON, supra note 2, at 104 (finding from survey results that wh collar workers); supra notes 130-35 and accompany U.K.). 158. See WILLIAMS ET AL., supra note 86, at 36.

M TRAVIS 12/16/2009 2:55:14 PM

412 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

1 just as the studi

uld like and many mothers working fewer hours than

159. See id. at 3, 31. In the U.S. in 2000, over 26 percent of men were working fifty or more hours per week, while less than 9 percent were working less than thirty hours per week. See JACOBS &

while long-hour full-time jobs are held predominantly by men.159 Researchers have linked this time divide to “a new occupational divide between jobs that demand excessively long days and jobs that provide neither sufficient time nor money to meet workers’ needs.”160 For those women working short-hour part-time positions, the negative consequences are severe and long-lasting in terms of downward occupational mobility, compensation and benefits, and promotion opportunities,16

es in Women and Employment have found in the U.K.162 As in the U.K., there is evidence that the growing bifurcation of U.S.

jobs into long-hour and short-hour positions is not a straightforward reflection of workers’ preferences.163 A growing majority of U.S. workers report a significant mismatch between their actual and ideal working hours.164 Specifically, “there is a growing time divide between those working especially long weeks, who would prefer to work less, and those working relatively short weeks, who would prefer to work more.”165 Because men are over-represented in the former group, while women are over-represented in the latter group, “[t]he result is many fathers working longer hours than they wo

they would like.”166 In the U.S., it appears that the work-family balance option that is most

frequently cited as “ideal” by workers who are unhappy with their number

GERSON, supra note 2, at 33, fig. 1.4A. During the same year, less than 12 percent of women were working fifty or more hours per week, while nearly 20 percent were working for thirty hours or less. See id. at 33, fig. 1.4B, 35. 160. See JACOBS & GERSON, supra note 2, at 8 (emphasis omitted). 161. See WILLIAMS, supra note 83, at 96-100; WILLIAMS ET AL., supra note 86, at 30-31. 162. See Crompton & Lyonette, supra note 38, at 220-22; Dex et al., supra note 31, at 69; Fagan et al., supra note 38, at 202; Harkness, supra note 39, at 249-50; Himmelweit, supra note 41, at 349, 355; McRae, supra note 33, at 180, 187-93; Rubery, supra note 41, at 295, 306. 163. See JACOBS & GERSON, supra note 2, at 5 (concluding from a national survey of over 3000 workers in the 1990s that “workers’ actual time at work does not necessarily reflect their desires”). 164. See id. at 64-67, tbl. 3.1, 77 (reporting data from 1997 and concluding that “most American workers experience a significant gap between how much they work and how much they would like to work,” and that this group is growing over time); see also WILLIAMS ET AL., supra note 86, at 30 (describing a national survey of 500 dual-career families in which 65 percent of women in full-time jobs reported wanting to work part-time). In a 1997 survey, only one-fifth of workers surveyed reported that their actual working hours matched their ideal working hours, while three-fifths reported that their actual hours were longer than their ideal and one-fifth reported that their actual hours were shorter than their ideal. See JACOBS & GERSON, supra note 2, at 64, tbl. 3.1. These figures do not include the unemployed, who should be added to those who desire greater working hours. See id. at 64. 165. JACOBS & GERSON, supra note 2, at 5, 36, 60, 63 (drawing this conclusion from the results of a national survey of over 3000 workers during the 1990s); see also Schultz & Hoffman, supra note 8, at 132. 166. See WILLIAMS ET AL., supra note 86, at 3 (concluding that “[i]nflexible, all-or-nothing workplaces drive women out of breadwinner roles and men out of caregiver roles”).

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 413

y few workers are regu

hours per week for men, and from

hey are remarkably united in their aspirations . . . [a]cross gender, class, and family

er week,” and noting that “[t]he United States has relatively

reased,

ercent of men and less than 9 percent of o

71-76 (describing various institutional isions).

of working hours is to combine caregiving with approximately thirty to forty hours of paid work per week,167 which mirrors the findings in the U.K. This aspiration is one thing that unites workers across both gender and class lines,168 and the labor market is not producing sufficient employment options to meet this particular preference.169 Despite workers’ reported “ideal,” the forty-hour workweek has become increasingly less typical since 1970 for both women and men,170 and ver

larly working thirty-five to forty hours per week.171 Just like the European researchers in Women and Employment,

American researchers studying this data have seriously questioned the preference theory for explaining the class-based and gendered nature of work-family decisions.172 Employees’ own reports of a growing divide between actual and ideal working hours have been bolstered by research identifying a variety of social and institutional factors that shape, limit, and constrain work-family decisions. For example, the rise in dual-earner and single-parent families can increase the time demands felt by many workers, even when the average time that individuals spend at work remains

167. See JACOBS & GERSON, supra note 2, at 64-69; WILLIAMS ET AL., supra note 86, at 30. In a large-scale 1997 survey, respondents were grouped into categories by their current number of working hours per week: those working less than thirty hours, those working thirty to thirty-nine hours, those working forty to forty-nine hours, those working fifty to fifty-nine hours, and those working sixty or more hours. See JACOBS & GERSON, supra note 2, at 66, fig. 3.1. The average number of hours reported as “ideal” within each of these categories was surprisingly similar across categories and for both men and women: ranging from approximately thirty-two to forty-two approximately twenty-seven to forty-one hours for women. See id. 168. See JABOBS & GERSON, supra note 2, at 78 (“While workers are divided in the options and constraints, t situation.”). 169. See id. at 77; see also WILLIAMS ET AL., supra note 86, at 30 (emphasizing the “inability to find good jobs requiring 30 to 40 hours p few good, 35 to 40 hour per week jobs”). 170. See JACOBS & GERSON, supra note 2, at 32-33, fig. 1.4 (reporting a 10 percent decline in the percentage of the workforce working a forty-hour workweek between 1970 and 2000, and noting that although the average workweek remained stable, “[v]ariation around the average has inc marking the emergence of both longer and shorter workweeks for different groups of workers”). 171. See WILLIAMS ET AL., supra note 86, at 30 (citing a study finding that among employed workers between the ages of twenty-five and fifty, only 3 p w men work between thirty-five and forty hours per week). 172. See JACOBS & GERSON, supra note 2, at 36-37 (concluding that “it is neither possible nor convincing to attribute the growing time divide among American workers simply to purely personal, private preferences,” or “to matters of personal taste and choice”); see also Kessler, supra note 85, at 442 (criticizing rational choice theory for assuming that “women’s cultural caregiving is a mere choice, for which the state owes no support and employers owe no accommodation”); Michael Selmi, Family Leave and the Gender Wage Gap, 78 N.C. L. REV. 707, 736-42 (2000) [hereinafter Selmi, Family Leave] (arguing that “until women are afforded the same opportunities as men and then choose differently from men, we should not blindly conclude that the observed disparities [in the labor market] are the product of choice”); Williams, supra note 85, at 14 arrangements that constrain women’s work-family dec

M TRAVIS 12/16/2009 2:55:14 PM

414 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

which would be essential to truly “test” the preference hypo

r market participation and some directions that future research might take.

173. See JACOBS & GERSON, supra note 2, at 42.

stable.173 In the face of these rising demands, many women are still experiencing widespread employment discrimination, hostile work environments, and negative stereotyping in their status as mothers, which further contributes to the push of women out of the labor market after having children.174 Several legal scholars have analyzed the ways in which these “pushes” are compounded by a set of legal regulations that create incentives for employers to divide the labor force into long-hour, full-time salaried positions that are exempt from overtime obligations, and short- hour part-time positions that avoid the costs of various employment benefits.175 And of course, this list must include the lack of a full-day school schedule and of high-quality, affordable childcare options, the availability of

thesis.176 Overall, the data on constrained decisionmaking in Women and

Employment (in combination with the older but largely corroborating evidence from American work-family research) calls the preference theory into sufficient question that consideration of a “choice model” for framing future work-family initiatives becomes warranted. Part B therefore returns to the question of whether and how a choice model might be translated into meaningful policy objectives. It also highlights some of the major challenges that exist in attempting to expand the range of viable options for combining care work with labo

B. Legal and Policy Implications

The last four articles in Women and Employment provide a wide range of proposals for new work-family regulations in the U.K. In one of these

174. See WILLIAMS ET AL., supra note 86, at 43-47 (providing a taxonomy of the various types of “maternal wall” bias that pregnant women and mothers face in the workplace and documenting the rise in “family responsibilities discrimination” lawsuits). 175. See, e.g., JACOBS & GERSON, supra note 2, at 36-37, 163-64 (describing the effects of the Fair Labor Standards Act’s white collar exemptions from overtime premiums and other forms of legal regulation that contribute to the “occupational divide”); Schultz & Hoffman, supra note 8, at 138-39 (describing aspects of current overtime and benefits laws that create incentives for employers to bifurcate their workforces into employees with overly long and overly short hours, including the Fair Labor Standards Act’s white collar exemptions from overtime premiums and the Employee Retirement Income Security Act’s failure to mandate benefits for part-time workers). 176. See Selmi & Cahn, supra note 51, at 24-25 (arguing that restructuring the school day and providing publicly-financed, full-day preschool and daycare are necessary to “allow[] lower-income women to make more informed and constructive choices concerning work-family issues because they facilitate both workforce attachment and increased income”); cf. Himmelweit, supra note 41, at 350-52 (describing the unique challenges to the “commodification of care”).

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 415

that a focus on choice might bring

jobs with a thirty-five-hour standard workweek.184 This narrower objective

, supra note 20, at 276.

uding that there is “no ‘one size fits all’ ilemmas of contemporary Americans”).

i, supra note 3, at 575-76.

see also JACOBS & GERSON, supra

articles, Professor Jane Lewis attempts to translate her “genuine choice” model into a concrete policy agenda.177 “Real or genuine choice to engage in paid and unpaid work,” says Lewis, “requires a policy package that addresses the possibility of first, being able to care, which requires care leaves that are paid, and/or reduced or more flexible working hours; and, second, the possibility of being able to work, which requires the provision of care services and/or money to buy services.”178 Leading American sociologists Jerry A. Jacobs and Kathleen Gerson similarly conclude that providing workers with “genuine choices” on work-family balance would require “a range of policies that enable people to meet their work and family aspirations, as they define them now and as they change over the course of their lives,” including, among other things, community-based and on-site childcare, comprehensive after-school programs, income supports to allow low-income workers the option of working less, stronger equal opportunity policies that encompass parental status, and greater workplace flexibility.179 “Creating a ‘culture of tolerance,’ that recognizes the diversity of needs among parents and workers,” suggest Jacobs and Gerson, “may be the most important contribution”

to the national work-family debate.180 While entirely accurate and highly desirable, calls for such broad-

reaching policy packages likely contribute to Professor Selmi’s concern about a choice model premised on “a desire to support all choices for women.”181 As Selmi suggests, such a desire, while laudable, risks opposition by being mischaracterized as asserting, “that women ought to be able to do whatever they want without penalty.”182 Within their comprehensive policy statements, however, the leading voices for “choice” (both in the U.K. and the U.S.) highlight the particular importance of regulating working time: specifically, the importance of “reducing time for the overworked and providing more work to the underemployed.”183 Other leading American work-family scholars have placed primary emphasis on this objective as well, urging activists to join the international labor movement and focus legal reforms on expanding the availability of quality

177. See Lewis 178. Id. 179. JACOBS & GERSON, supra note 2, at 170-80 (concl solution to the time d 180. Id. at 170. 181. See Selm 182. See id. 183. See JACOBS & GERSON, supra note 2, at 170; see also Lewis, supra note 20, at 280-81. 184. See, e.g., Schultz & Hoffman, supra note 8, at 132, 140-41;

M TRAVIS 12/16/2009 2:55:14 PM

416 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

note 2, at 125 (summarizing reduced-hour legislation in Europe).

could provide a viable focus for a new “choice model” to move forward the work-family debate. As explained above, the empirical data in both the U.K. and the U.S. suggests that the labor market is not supplying a sufficient number of thirty- to forty-hour-per-week jobs to meet the demand expressed by both full-time and part-time workers across both gender and class lines. Thus, seeking regulatory methods to expand the supply of such jobs indeed could be framed effectively in terms of expanding “choice.”

While such a focus could provide useful empirical grounding for a new “choice model,” the question remains whether such a perspective could retain a normative grounding in gender equality. In other words, could feminist legal scholars still rely on the persuasive power of a gender equality norm as the basis for advocating such legal reform? Or would shifting away from an “equal labor market results” perspective or an “equal valuation of care work” perspective to this type of “choice” perspective require a corresponding shift away from gender equality as a central theme in the work-family policy arena? Put another way, is a choice perspective in this context better characterized as an “equal choice” or a “genuine choice” model?

The fact that thirty- to forty-hour-per-week jobs are under-supplied for both women and men presents a challenge to framing expansion of such an option in gender equality terms.185 However, the consequences of the bifurcation of working hours are markedly gendered. Because the growing dispersion of working hours results in a disproportionate number of women in short-hour part-time positions and a disproportionate number of men in long-hour full-time positions, the economic consequences are much more severe for women, in the aggregate, than for men. As noted above, short- hour part-time positions cause long-lasting reductions in compensation levels, available benefits, training and advancement opportunities, and occupational mobility.186 As a result, more women than men end up in positions of economic dependency due to the inability to secure full-time jobs with reasonable hours. Conversely, men face more severe consequences on the care side of the equation, as even more fathers than mothers report an inability to spend their desired amount of time with their

185. See Selmi & Cahn, supra note 51, at 13-14 (arguing that because “few individuals, men or women, have the choice to pursue a meaningful career while working reduced hours, . . . it is not at all clear why creating such an option should be a principal policy focus”); see also Selmi, supra note 3, at 576 (“From the equality perspective, the primary question is whether women are afforded the same, or equivalent, choices as men.”). 186. See supra notes 161-62 and accompanying text.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 417

families.187 A gender equality frame is bolstered not only by these gendered

effects, but also by the mediating role that employment discrimination plays between the under-supply of thirty- to forty-hour-per-week jobs and the gendered bifurcation of working hours. In the current all-or-nothing work environment, women face an added “push” towards less labor market attachment in the form of pervasive hostility and negative competence stereotypes about mothers.188 To the extent that men face employment discrimination in their role as fathers, that factor typically pushes in the opposite direction of increasing labor market attachment. For example, when employers interfere with, deny, or retaliate against fathers taking family leave based on the stereotypic assumption that men’s wives can and should be providing the care, men are often pushed away from engagement in caregiving activities.189

Thus, it is possible to tether a choice model to gender equality ideals, particularly with a narrow focus on expanding the supply of full-time jobs with reasonable hours. The expansion of this particular “choice” has the potential both to help women obtain workplace equality with men and to help men obtain caregiving equality with women. This potential, however, will face some significant challenges as advocates attempt to translate a “choice” ideal – even one grounded in well-established gender equality norms – into concrete policy proposals. In Lewis’s article in Women and

187. See WILLIAMS ET AL., supra note 86, at 28; cf. Michael Selmi, Care, Work, and the Road to Equality: A Commentary on Fineman and Williams, 76 CHI.-KENT L. REV. 1557, 1566 (2001) [hereinafter Selmi, Care, Work] (arguing “that women are entitled to the same choices that men have,” but recognizing that, “[g]iven the persistence of gendered parenting patterns, women have, as a practical matter, a different array of choices than men,” and “that women are the ones who make the trade-offs” when balancing work and care). 188. See WILLIAMS, supra note 83, at 37-39 (arguing that a system that allows women only two alternatives – either “perform as ideal workers without the flow of family work and other privileges male ideal workers enjoy,” or “take dead-end mommy-track jobs or ‘women’s work’” – is a system “that discriminates against women”); WILLIAMS ET AL., supra note 86, at 43-49 (documenting how mothers’ work-family decisions “often occur within the context of family responsibilities discrimination,” and describing a range of “maternal wall” biases that push women out of the workforce, including open hostility towards mothers, benevolent sexism, assumptions of role incongruity between working and mothering, negative competence assumptions, and attribution biases). 189. See Joan Williams & Stephanie Bornstein, The Evolution of “FRED”: Family Responsibilities Discrimination and Developments in the Law of Stereotyping and Implicit Bias, 59 HAST. L.J. 1311, 1311-13, 1320-21 (2008) (documenting the specific types of family responsibilities discrimination against men); see also Martin H. Malin, Fathers and Parental Leave Revisited, 19 N. ILL. U. L. REV. 25, 39-42, 55 (1998) (summarizing studies documenting workplace hostility that deters men from using family leave policies, described as “the ‘your wife should do it’ syndrome,” and concluding that “[r]educing workplace barriers is crucial to increasing fathers’ involvement with their children”); WILLIAMS ET AL., supra note 86, at 49 (explaining that when women reduce their labor force attachment “because their husbands encounter the expectation that they should not take leave because ‘your wife should do it,’ they are not freely opting out – they are being pushed out by gender discrimination”).

M TRAVIS 12/16/2009 2:55:14 PM

418 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

Employment, she describes some of these potential limits to a model grounded in “choice.”

1. The Less-Is-More Choice Paradox and Other Challenges

Although advocates often describe a choice perspective as more democratic,190 using regulation to make some, but not all, work-family combinations more readily available is inevitably normative.191 It is obviously value-laden, for example, to regulate working hours to allow more workers to choose a work-family combination with a thirty-five hour workweek, rather than to provide cash for care to allow more workers to choose exclusive caregiving. Thus, to the extent that a choice perspective is used to advance anything other than “all choices for women,” selecting which options to expand the availability of may end up just reviving the divides between supporters of the “equal labor market results” and the “equal valuation of care work” perspectives – which is precisely what a choice perspective hoped to avoid. However, this is where the empirical evidence in Women and Employment may help, by providing a rational basis for identifying the most under-supplied option as a uniting focal point.

Another challenge facing a choice perspective is the risk that policy- makers may view “choice” as the absence of regulatory intervention.192 Such an approach would only expand workers’ choices if an unregulated labor market allows for the exercise of unconstrained preferences regarding work-family combinations. As the evidence in Women and Employment demonstrates, that is highly unlikely to be the case. As an example of this risk, Lewis highlights the conservative Danish government’s invocation of “men’s freedom to choose” as a basis for eliminating its “‘daddy leave’” policy, which had designated a portion of paid family leave solely to fathers on a “‘use it or lose it’” basis.193 Without such affirmative incentives for fathers to increase their engagement in care work, such deregulation actually constricted “choice” by pushing workers back to the default gendered division of labor that the existing social, institutional, and socioeconomic constraints facilitate and support.

190. See, e.g., Lewis, supra note 20, at 275. 191. See id. (explaining that “the meaning of ‘choice’ in this policy field is often value-laden,” particularly if government intervention is used only to expand one particular choice). 192. See id. 193. See id. at 275-76. The U.K.’s opt-out from the European Union’s Working Time Directive may be considered another example of this risk, see infra notes 204-07 and accompanying text, to the extent that the opt-out can be described as “in reality an absence of policy on working hours,” see Himmelweit, supra note 41, at 355.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 419

On the other hand, the mere expansion of options, “without changing the opportunity structures” of those options, is unlikely to fundamentally alter the choices that workers can make.194 In other words, making a wider range of options “more feasible and fair,” is more important than simply maximizing the options from which workers may choose.195 As Jacobs and Gerson argue, we need “policy approaches that include not just more choices but fundamental structural reforms.”196

Paradoxically, it is possible that options actually need to be restricted for choices to meaningfully expand, at least for women. This “less-is-more choice paradox” exists, in part, because work-family decisions are frequently made on a household basis, rather than individually. Specifically, many women are unable to “choose” increased labor market participation if their male partners are unwilling (or unable) to engage in more care work. As Lewis so aptly observes, “men’s choice not to care affects women’s choices on whether and how to combine paid and unpaid work.”197 This sentiment is shared by another author in Women and Employment, who further explains:

[G]iven existing gender norms, providing too much “choice” may work against gender equality; this is because choices that people make about caring responsibilities are made in the context of families where gender norms and power imbalances are often at their most acute. Thus, giving families choice may make equality less attainable for women than if more uniform good practice were to be encouraged.198

Of course, no form of state regulation can or should force men to provide more care, which would be unlikely to serve the best interests of the care recipients.199 But regulation can create incentives for, or remove disincentives against, men’s greater participation in caregiving. Such incentive structures are necessary to counteract or at least mitigate the existing social and institutional pressures that currently make it so costly for men to reduce their labor market attachment. Given that 95 percent of

194. See JACOBS & GERSON, supra note 2, at 180. 195. See id. at 181. 196. Id. 197. Lewis, supra note 20, at 276, 281-82 (arguing that “further enlargement of choices for women depends on men changing their behavior,” and observing that “there is little public recognition of the extent to which men’s choices affect those taken by women”); see also Selmi, Family Leave, supra note 172, at 742 (arguing that “equalizing choices between men and women may entail changing the choices currently available to both men and women”). 198. Himmelweit, supra note 41, at 364. 199. See id. at 364 (observing that “men cannot be forced to care, nor would attempting to do so be in the interests of those they care for (even if for centuries women have had little alternative”)); Lewis, supra note 20, at 276 (observing that “‘[c]ompulsion to care’ threatens the moral qualities of attentiveness, responsibility, competence and responsiveness identified by the feminist theorists of care”).

M TRAVIS 12/16/2009 2:55:14 PM

420 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

fathers in a recent study reported a desire to spend more time with their families,200 there is reason to suspect that changing the incentive structures regarding men’s work-family balance decisions may go a long way toward reducing the gendered division of both waged and unwaged work. This highlights the importance of policies like Denmark’s now-defunct “daddy leave,” and other government initiatives to encourage fathers to take on additional caregiving responsibilities.201

While such a regulatory strategy may help undermine the male career norm of uninterrupted work with unlimited hours,202 which contributes to the gendered bifurcation of working hours, it would do so only indirectly. A more direct route for expanding women’s ability to “choose” greater labor market attachment (or less, for those women currently in long-hour full-time jobs) would be to use regulation to expressly restrict the option of long-hour full-time work – i.e., to attack the male career norm itself.203 Restricting that option could expand the availability of full-time jobs with reasonable hours, which could meaningfully expand women’s work-family options.

In Women and Employment, several of the authors illustrate this less- is-more choice paradox with the U.K.’s treatment of the European Union’s 1993 Working Time Directive.204 This Directive requires member states to limit workers’ average workweek to a maximum of forty-eight hours and to ensure a daily rest period of eleven consecutive hours in each twenty-four hour period.205 In response to demands by the U.K., a deal was struck to

200. See WILLIAMS ET AL., supra note 86, at 28 (describing the study). 201. See Lewis, supra note 20, at 275-77 (describing Denmark’s experience and concluding that state policies that “make it possible for men to continue to avoid taking care responsibilities” will inevitably mean “that women continue to be disadvantaged in the labour market’); cf. Martin H. Malin, Fathers and Parental Leave, 72 TEX. L. REV. 1047, 1048-49 (1994) (arguing that “most men desire a greater role in child care but are precluded from it by significant workplace barriers,” and proposing “to expand the law’s focus to include the need for workplace accommodation of the family responsibilities of men”); Selmi, supra note 3, at 597 (emphasizing the importance of policies that “get men to act more like women,” rather than policies that enable “women to spend more time out of the workplace”); Selmi, Care, Work, supra note 187, at 1558 (arguing that “an important component to obtaining greater gender equality lies in changing the labor force patterns of men by getting them to shoulder more of the burden of home work”); Selmi, Family Leave, supra note 172, at 708, 712 (arguing that “if there is to be greater equality for women in the workplace, it will be necessary for men to change their behavior,” and proposing amendments to the Family and Medical Leave Act “to create greater incentives for men to take leave around the birth or adoption of a child”). 202. See Lewis, supra note 20, at 277 (describing research in Norway finding that “state initiatives on parental leave to encourage fathers to take care of young children are crucial to changing workplace culture”). 203. See id. at 276 (arguing that “further enlargement of choices for women depends on . . . changes to the male career norm”). 204. See Deakin & McLaughlin, supra note 41, at 324; Himmelweit, supra note 41, at 354-55; Lewis, supra note 20, at 280; Rubery, supra note 41, at 305. 205. Council Directive 93/104, Concerning Certain Aspects of the Organization of Working Time,

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 421

include a provision allowing member states to adopt an “opt-out” to the forty-eight hour maximum workweek when enacting the Directive into their national legislation.206 The U.K. is the only member state that has incorporated a broad opt-out provision, which allows most adult workers in the U.K. to voluntarily sign an individual agreement with their employer stating that they are willing to work more than forty-eight hours per week.207

In enacting and defending the opt-out against fierce and ongoing opposition, the U.K. government has invoked choice rhetoric at every turn. A government website even provides employers with a sample opt-out agreement.208 Given the highly problematic nature of gauging voluntariness within the inherent power imbalance in many employment relationships (particularly during a recession), it is not surprising that an estimated one- third of British workers have signed an opt-out agreement,209 and approximately 16 percent (or four million workers) are regularly working more than forty-eight hours per week.210

Although the law makes it illegal for an employer to discriminate against a worker for refusing to sign an opt-out,211 research suggests that many individuals do not perceive any “choice” about opting out, and the anti-retaliation law is rarely enforced.212 One-third of workers who have signed opt-outs report having been required to do so, and two-thirds of workers in the U.K. are unaware that the forty-eight hour workweek limit even exists.213 In one survey, 65 percent of employers that responded

1993 O.J. (L 307) 18-24 (EC), available at <http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri= CELEX:31993L0104:EN:HTML> (last visited Oct. 31, 2009). 206. See id., art. 18, available at <http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri= CELEX:31993L0104:EN:HTML; see also European Commission, Frequently Asked Questions about Working Time, <http://ec.europa.eu/bulgaria/abc/faq/working_time_en.htm> (last visited Oct. 31, 2009). 207. The U.K. government hosts a website called “Directgov” that provides information to its citizens, including a description of the opt-out provision of the E.U. Working Time Directive. See Directgov, Working Time Limits (the 48 Hour Week), <http://www.direct.gov.uk/en/Employment/ Employees/WorkingHoursAndTimeOff/DG_10029426 (last visited Oct. 12, 2009). 208. See id. 209. See EU Working Time Directive: Advice and Support, The European Union Working Time Directive, <http://www.eu-working-directive.co.uk/> (last visited Oct. 31, 2009); European Commission, supra note 206. 210. See European Commission, supra note 206; European Trade Union Confederation: The Voice of European Workers, ETUC: Background Information to the Working Time Directive (Mar. 29, 2004), <http://www.etuc.org/a/551> (last visited Oct. 31, 2009). 211. See European Commission, supra note 206. 212. See Trades Union Congress: Britain at Work, Ending 48 Hour Working Week Opt-Out Is Affordable and Good for Social Stability (Dec. 12, 2008), <http://www.tuc.org.uk/work_life/tuc-15747- f0.cfm> (concluding that “[w]hat the Government describes as an individual opt-out is in reality no choice at all for many long-hours workers” (internal quote omitted)). 213. See European Trade Union Confederation: The Voice of European Workers, supra note 210;

M TRAVIS 12/16/2009 2:55:14 PM

422 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

reported having asked their employees to sign an opt-out.214 In most of those companies, a majority of the workers had signed the agreement, and in 28 percent of those companies not a single worker had refused to sign.215 Many employers have adopted a general policy of presenting newly-hired workers with an opt-out agreement along with their employment contract, which the European Commission has concluded “de facto undermines the freedom of choice of the worker.”216 The conclusion that opt-outs do not necessarily reflect workers’ unconstrained preferences is further bolstered by the fact that over 70 percent of salaried workers performing unpaid overtime, and nearly 57 percent of hourly workers performing paid overtime, report a desire to reduce their working hours.217

Even if one could be assured that all opt-out agreements reflect those workers’ preferences to work long hours, several authors in Women and Employment argue that the availability of such an option restricts the choices available to others, particularly women in dual-earner households. Economics Professor Susan Himmelweit observes that although the U.K. government defends the opt-out “on the grounds of individual choice,” the government has failed to recognize “that one worker’s ‘choice’ to work long hours may result in their partner’s choices with respect to employment being heavily curtailed.”218 When a very long-hour work culture is combined with the large gender pay gap and expensive childcare in the U.K., the logical financial decision is for women to reduce their working hours and for men to increase them.219 Allowing the “choice” of working very long hours thus effectively restricts the “choice” for both parents of young children to work full-time.220

Because the opt-out allows employers to retain the male career norm that is built around a culture of very long hours,221 even full-time workers

see also Workers’ Liberty for International Working Class Solidarity and Socialism, The Long Hours Scandal, <http://www.workersliberty.org/node/4460> (July 21, 2005) (suggesting that one-third of workers who have signed an opt-out agreement were pressured into doing so by their employers). 214. See European Commission, supra note 206. 215. See id. 216. See id. 217. See Trades Union Congress: Britain at Work, supra note 212; see also Fagan et al., supra note 38, at 208 (finding that most of the full-time managerial and professional women in the U.K. desire to reduce their working hours). 218. Himmelweit, supra note 41, at 354; see also Lewis, supra note 20, at 280 (explaining that the significance of the opt-out to the E.U.’s Working Time Directive is that “British fathers have among the longest working hours in Europe, which constrains the time they can spend on care work and the time their partners can spend on paid work”). 219. See Himmelweit, supra note 41, at 354. 220. See id. 221. See Deakin & McLaughlin, supra note 41, at 324 (arguing that the E.U.’s Working Time Directive “had little impact in changing prevailing organisational practices” because of “the ease with

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 423

them

who are not regularly working over the forty-eight hour maximum end up working relatively long workweeks. Full-time workers in the U.K. average forty-four hours per week, which is the highest of all E.U. member states.222 The U.K. is the only E.U. member state in which the average workweek has increased since the enactment of the Working Time Directive.223 More significantly, the opt-out has contributed to a more severe bifurcation of working hours between women and men in the U.K. than in any other European country.224 The U.K. has the highest percentage of workers performing very long hours, and a very high percentage of workers in short-hour part-time positions.225 Because the long-hour jobs are performed disproportionately by men, fathers in the U.K. work the longest hours in Europe.226 In contrast, British women have a lower rate of full-time work than do women in other European countries, with nearly half of the U.K.’s female workforce employed part-time and a majority of mothers with young children working very short hours.227 Himmelweit concludes, therefore, that simply attempting to “maximi[ze] individual freedom” by “making working hours a matter of individual choice,” ends up doing “nothing to challenge gender inequalities and, indeed, reinforces

.”228 Thus, to the extent that a choice perspective is grounded in gender

which employers could impose opt-outs on individual workers”); Rubery, supra note 41, at 305 (“The abolition of the voluntary opt out from UK working time regulations is the first and most important policy measure that needs to be taken to tackle the long-hours culture.”). 222. See European Trade Union Confederation: The Voice of European Workers, supra note 210. 223. See id.; see also European Commission, supra note 206. 224. See Himmelweit, supra note 41, at 354. 225. See id. 226. See id. at 349 (noting the “disproportionately male full-time employment”); Lewis, supra note 20, at 280 (observing that “British fathers have among the longest working hours in Europe”). 227. See Himmelweit, supra note 41, at 349 (describing the “large sector of disproportionately female relatively short hours’ part-time employment” and noting that a lower proportion of women in the U.K. work full-time than in the rest of Europe); Lewis, supra note 20, at 271 (noting that the majority of women with young children in the U.K. work short part-time hours); McRae, supra note 33, at 179-80 (observing a “steady rise in mothers’ part-time employment, from 19 per cent of all new mothers in 1979 to 54 per cent in 2002,” and noting that almost one-half of the entire female workforce is employed part-time); Scott, supra note 20, at 160-62 (noting that although women’s labor force participation rates are similar in the U.K. and the U.S., women have much higher rates of part-time employment in the U.K., with 42 percent of women in the U.K. labor force in 2002 working part-time). 228. Himmelweit, supra note 41, at 355, 364. In her article in Women and Employment, Himmelweit provides another example of the less-is-more choice paradox when describing government options for regulating childcare. See id. at 364-65. On one hand, “the policies that provide most choice to those looking for substitute care may have deleterious effect on gender inequalities through the employment conditions of those who provide that care,” she explains. Id. at 365. This is because paid caregivers can be “particularly vulnerable workers,” and they are virtually all women. Id. On the other hand, “[p]ublic provision or regulation of private provision may indeed restrict choice,” observes Himmelweit, “but that may be necessary to prevent the expansion of paid care entrenching or even worsening existing gender inequalities.” Id.

M TRAVIS 12/16/2009 2:55:14 PM

424 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

and Norway have generous “take-it-or- lose-

equality, the less-is-more choice paradox must be considered when translating “choice” into policy. In some circumstances, choice may be a zero-sum game, in which the expansion of choice for some (particularly men) may result in the restriction of choice for others (particularly women), and vice versa. In acknowledging this reality, Himmelweit ultimately concludes that equality actually would best be served through greater homogeneity in working hours – specifically, by expanding workers’ ability to choose the particular option of a full-time job with reasonable hours.229 She notes that countries reporting the lowest levels of work- family conflict have the most similarity between women and men in the distribution of both working hours and caregiving work.230 Mothers and fathers in Scandinavia, for example, report the lowest levels of work-family conflict, the highest levels of work-family fit, and no meaningful gender differences on either dimension.231 In Scandinavia, most men and women work similar hours in relatively short-hour full-time jobs (with a relatively small gender pay gap),232 and almost two-thirds of fathers perform at least half of their household’s childcare responsibilities.233 Unlike the U.K., the Scandinavian countries do not permit opt-outs from the E.U.’s Working Time Directive, and both Sweden

it” paternity leave benefits.234 Like Himmelweit, American sociologists Jacobs and Gerson similarly

find their way from a starting point of “genuine choices” to an endpoint of “a more even distribution of working time across jobs.”235 A more even distribution in working hours may be achieved not only with regulation that restricts long-hour workweeks, which is the primary emphasis of the authors in Women and Employment, but also with regulation that makes

229. See id. at 355, 364. 230. See id. 231. See E. Jeffrey Hill et al., Studying “Working Fathers”: Comparing Fathers’ and Mothers’ Work-Family Conflict, Fit, and Adaptive Strategies in a Global High-Tech Company, 1 FATHERING: A JOURNAL OF THEORY, RESEARCH, AND PRACTICE ABOUT MEN AS FATHERS 239, 255-57 (2003), available at <http://findarticles.com/p/articles/mi_m0PAV/is_3_1/ai_111268932/?tag=content;col1> (last viewed Oct. 31, 2009). These findings came from “the largest known single-issue work-life survey,” which gathered nearly 26,000 responses from IBM employees in forty-eight countries in 2001. Id. at 246. The study defined “work-family conflict” as role incompatibility between work and family domains, id. at 244, and it defined “work-family fit” as a person’s “overall assessment of how well she or he has been able to integrate paid work and family life,” id. at 245. 232. See Himmelweit, supra note 41, at 355 (observing that in Scandinavia, “full-time working hours are much shorter than in the UK but both men and women tend to work full-time,” and “there is greater equality between men and women in the distribution of hours across paid and unpaid work, less of a gender pay gap and greater sharing of caring roles than in the UK”). 233. See Hill et al., supra note 231, at 255-57. 234. See id.; Himmelweit, supra note 41, at 355. 235. See JACOBS & GERSON, supra note 2, at 183.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 425

directly to the preferences (and needs) of a great many workers indeed.238

lation might differ from

short-hour part-time positions less attractive for employers, such as requiring pro rata pay and benefits.236 By addressing the growing bifurcation of working hours and increasing the availability of full-time jobs with reasonable hours, these policies would help promote gender and class equality both in the workplace and in the home.237 Such an endpoint – one that arguably may be characterized as restricting some options and seeking greater homogeneity in work-family balance approaches – does not necessarily reflect an abandonment of a choice ideal. The evidence on constrained choice described above suggests just the opposite: that regulating work to increase the supply of full-time jobs with reasonable hours speaks most

2. The Role of Reflexive Law in Regulating Choice

In Lewis’s article in Women and Employment, she distinguishes a choice model of equality from the two other dominant equality perspectives – equal labor market results and equal value for care work – as being process-based rather than outcomes-oriented.239 “[I]t may be strategic for those concerned about gender equality,” she suggests, “to focus on inputs” that maximize workers’ “‘real agency freedom’” in making work-family decisions, rather than trying to get the government to prioritize either of the two substantive definitions of equality within its work-family policies.240 While this distinction is an intriguing one, Lewis does not further develop it in her article to demonstrate how inputs-focused regu

or improve upon an outcomes-based legal regime. In contrast, the authors of another article in Women and Employment

provide extensive discussion of how process-based regulation – what they refer to as “reflexive law” – differs from outcomes-based regulation – what they refer to as “‘command-and-control’” regulation or “prescriptive” law.241 These authors, Cambridge Law Professor Simon Deakin and Cambridge Research Associate Colm McLaughlin, analyze in particular the potential role and limitations of a reflexive law approach in achieving gender pay equity in the U.K.242 Again, while the distinction between reflexive and prescriptive law is an intriguing one, the authors do not

236. See id.; Himmelweit, supra note 41, at 354-55. 237. See JACOBS & GERSON, supra note 2, at 183. 238. See id. 239. See Lewis, supra note 20, at 274-76. 240. Id. at 275-76 (emphasis added). 241. Deakin & McLaughlin, supra note 41, at 319-21. 242. See id. at 322-26.

M TRAVIS 12/16/2009 2:55:14 PM

426 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

o of the most signi

el and t

consider the role of reflexive law in the work-family arena beyond the narrow issue of compensation, and they do not consider what role, if any, reflexive law might play in advancing a choice model of equality. Thus, the authors of these two articles miss an opportunity to connect tw

ficant emerging concepts within the work-family debate. The same two concepts have been emerging in similar parallel fashion

within the American work-family literature. As Professor Selmi has observed, a growing group of work-family scholars has been attempting to shift the gender equality frame to one emphasizing increased choice.243 At the same time, a growing group of employment law scholars has been shifting focus from external regulatory standards to a “new governance” approach that emphasizes public oversight of private, self-regulatory initiatives.244 The two articles in Women and Employment that highlight these tandem developments provide an implicit call to these two groups to begin a more explicit dialogue that considers, more fully, the role that reflexive law might play in advancing a choice model of equality. To the extent that a choice perspective might be meaningfully linked to a new governance lawmaking approach, such efforts could provide the theoretical foundation that Professor Selmi currently finds lacking in a choice mod

hereby help avert the policy “stalemate” that he predicts will result. The Deakin and McLaughlin article provides a valuable starting

point in considering this new work-family research agenda. The authors begin by distinguishing reflexive regulation both from “‘command-and- control’” regulation, which uses prescriptive rules with civil or penal sanctions for non-compliance, and from deregulation, which would move

243. See Selmi, supra note 3, at 575-76. 244. See Rachel Arnow-Richman, Public Law and Private Process: Toward an Incentivized Organizational Justice Model of Equal Employment Quality for Caregivers, 2007 UTAH L. REV. 25, 63- 74 (describing this trend in employment discrimination scholarship); Cynthia Estlund, Rebuilding the Law of the Workplace in an Era of Self-Regulation, 105 COLUM. L. REV. 319, 377-83 (2005) (proposing a system of monitored employer self-regulation to enforce labor standards); Orly Lobel, Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought, 89 MINN. L. REV. 342, 344-45 (2004) (describing the growth of the “new governance model” as a “more participatory and collaborative model,” in which “[l]awmaking shifts from a top-down, command-and-control framework to a reflexive approach, which is process oriented”); Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 COLUM. L. REV. 458, 553-68 (2001) (advocating a regulatory approach that creates incentives for employers to adopt structural problem-solving methods to reduce discrimination); Michelle A. Travis, Lashing Back at the ADA Backlash: How the Americans With Disabilities Act Benefits Americans Without Disabilities, 76 TENN. L. REV. 311, 356 n.273 (2009) [hereinafter Travis, Lashing Back] (describing “the recent scholarly trend to promote the value of process-based solutions for reducing employment discrimination and advancing worker rights”); Michelle A. Travis, Employment Protection for Atypical Workers: Proceedings of the 2006 Annual Meeting, Association of American Law Schools Section on Labor Relations and Employment Law, 10 EMPLOYEE RTS. & EMP. POL’Y J. 233, 266-69 (2006) [hereinafter Travis, Employment Protection] (describing the U.K.’s legislation giving workers a right to request part-time work as an example of the “shift in codifying rights in terms of process, rather than substantive outcomes”).

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 427

omic effic

of mandatory pay audits and information disclosure rules

away from statutory control back to “individual freedom of contract” or “market-based governance.”245 Reflexive law, in contrast, uses procedural norms to facilitate self-regulation to meet a particular policy aim.246 By definition, reflexive law permits and validates a range of solutions to a policy goal, and it attempts to promote institutional learning structures to develop “‘what works best’” within each organization.247 In the employment context, the overarching aim of reflexive law would be finding ways of “embedding the equality principle” into organizational practices and norms.248 American researchers have described the “new governance” approach in very similar terms, and have heralded it as a viable solution both to regulatory and market failures by recognizing “that econ

iency and democratic legitimacy can be mutually reinforcing.”249 Deakin and McLaughlin then consider examples of reflexive law that

the U.K. government has adopted to address the gender pay gap that traditional antidiscrimination law has been unable to eliminate.250 During the last decade, the U.K. government has established a variety of incentives to encourage institutional commitment to gender pay equity, including an award system to recognize firms that take exceptional steps in addressing equal pay, agency assistance for firms to voluntarily conduct equal pay reviews of their workforces, and an equal pay questionnaire that allows individuals who believe they are receiving unequal pay to request certain information from their employers to assess their relative compensation.251 In surveying the empirical evidence, the authors conclude that these approaches have had limited success, particularly in persuading employers to perform pay reviews.252 They advocate for a stronger form of reflexive law – but one that remains uncommitted to any particular substantive outcome – in the form

.253 In the absence of widespread governmental monitoring, however, the

evidence suggests that even those stronger forms of reflexive law are unlikely to succeed unless unions or other employee organizations jointly manage the process.254 The authors therefore conclude that one of the

245. Deakin & McLaughlin, supra note 41, at 319. 246. Id. at 319-20. 247. Id. at 320. 248. Id. at 326. 249. See Lobel, supra note 244, at 344. 250. See Deakin & McLaughlin, supra note 41, at 316-18, 322-26. 251. See id. at 317-18. 252. See id. at 322-23. 253. See id. at 322-25. 254. See id. at 323.

M TRAVIS 12/16/2009 2:55:14 PM

428 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

with deep skepticism about the effec

preconditions for effective reflexive laws is the existence of “bridging institutions,” which they describe as extra-legal structures for workplace deliberation and employee representation.255 Noting the detrimental effects that the U.K.’s opt-out from the E.U.’s Working Time Directive has had on gender equality, the authors further emphasize the importance of deliberative mechanisms at the workplace level rather than the individual level, such as collective bargaining or employee consultation groups.256 The absence of this institutional prerequisite in most workplaces in the U.K. (as in the U.S.) leaves the authors

tiveness of reflexive law strategies in this arena.257 The next step for American work-family scholars is to consider what

these insights might mean for the role of reflexive law within a broader work-family policy agenda, and specifically whether a reflexive law focus could advance a choice model of gender equality. Very little of the new governance literature focuses on work-family issues, with a few notable exceptions.258 One is the thoughtful work by Professors Vicki Schultz and Allison Hoffman that surveys European experiments with legislative incentives, negotiated solutions, collective bargaining, and private industry initiatives to achieve a reduced-hour workweek.259 Similar to Deakin and Mclaughlin, these authors conclude that, in order to succeed in the U.S., anything other than private industry initiatives would require the creation of “a stronger structure to bolster representation of employees’ interests for purpose of designing and enforcing corporate compliance.”260 Another example of new governance scholarship with a work-family focus is, Professor Rachel Arnow-Richman’s envisioning work an “incentivized organizational justice model” for future work-family regulation.261 Rather than granting substantive accommodation rights to workers with caregiving responsibilities, her proposed model would grant procedural rights

255. See id. at 320-21, 324, 326 (describing the role of bridging institutions as “receive[ing] and translate[ing] reflexive legal norms in a way which makes their implementation effective”). 256. See id. at 320, 324-25 (describing one form of reflexive law as imposing a default rule that allows the parties to negotiate variances to the statutory norm, and describing the opt-out as a failed version of this form because of the lack of “collective routes” for negotiation and “the ease with which employers could impose opt-outs on individual workers”). 257. See id. at 323-26. 258. See, e.g., Arnow-Richman, supra note 244, at 45-84 (articulating an incentivized organizational justice model that could provide greater equality for workers with caregiving responsibilities); Schultz & Hoffman, supra note 8, at 144-49 (assessing the viability in the U.S. of using various European reflexive law methods to incentivize a reduced-hour workweek); Travis, Lashing Back, supra note 244, at 266-69 (describing the benefits of a process-based right rather than mandated substantive outcomes for increasing workplace flexibility). 259. Schultz & Hoffman, supra note 8, at 144-49. 260. See id. at 147. 261. See Arnow-Richman, supra note 244, at 56.

M TRAVIS 12/16/2009 2:55:14 PM

2009] FUTURE OF WORK-FAMILY POLICY 429

mean

n between new governance ideas and choice-based gender equality theory.

elations, and others, this book provides a useful tool for b

requiring employers to engage in a good-faith interactive process to ingfully discuss requests for leaves or flexible-hour arrangements.262 While both of these projects are exciting contributions to the work-

family debate, their discussions of new governance methods are occurring only in parallel to the discussions about the emerging “choice model” of gender equality. In distinguishing her organizational justice model from outcomes-oriented approaches (such as increasing leave benefits), Arnow- Richman notes that the latter lacks the ability “to address the divergent needs and circumstances of all caregivers,” which suggests a role for choice rhetoric within her procedural approach.263 However, her analysis moves away from a gender equality frame, focusing instead on a broader goal of employment “quality” for all working caregivers.264 While there are good reasons for such a move, work-family scholars who have been advancing an equal choice model should consider whether proposals such as Schultz and Hoffman’s or Arnow-Richman’s might be used to help more effectively translate “choice” into a meaningful policy agenda. This suggestion reflects one of the most important contributions of the collection of works in Women and Employment: the implicit recognition of the potential for moving work-family debates forward by considering more explicitly the intersectio

IV. CONCLUSION

The collection of empirical and policy articles in Women and Employment: Changing Lives and New Challenges is a valuable contribution to the growing international dialogue on work-family policy. The empirical studies, with their particular attention to issues of class and culture, help deepen our understanding of the highly complex factors and relationships affecting individual work-family decisions. As the book’s editors observe, “[t]he interface between social science and policy is not an easy one.”265 However, by merging insights from experts of many different disciplines, including sociology, demography, economics, human geography, industrial r

ridging that gap. This Review has used the emerging debate over a “choice” model of

gender equality to demonstrate one of the ways in which the data and

262. See id. at 56-62. 263. See id. at 45. 264. See id. at 27-29 (contrasting her procedural model to “equality-based initiatives”). 265. Scott et al., supra note 18, at 13.

M TRAVIS 12/16/2009 2:55:14 PM

430 EMPLOYEE RIGHTS AND EMPLOYMENT POLICY JOURNAL [Vol. 13:385

ences” of work-life decisions are shared more equally by all.

analysis in Women and Employment might contribute to American work- family conversations. Hopefully, it will also help disprove the widely-held view recently summed up by General Electric Co.’s former CEO, Jack Welch, who asserted, “‘[t]here’s no such thing as work-life balance[,]’” but only “‘work-life choices’” that come with “‘consequences.’”266 By helping unsettle the assumption that work-life decisions are a straightforward reflection of unconstrained “choices,” the authors in Women and Employment help contribute significantly to the ultimate goal of designing policies that will indeed enable workers to achieve work-life balance and ensure that the “consequ

266. Tuna & Lublin, supra note 1.

Copyright of Employee Rights & Employment Policy Journal is the property of Workplace Fairness and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.