Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Findings and Recommendations We began our examination of sex segre- gation in the workplace by describing why it is costly for both individuals and society. Sex segregation affects the lives of many Americans. Women of all racial and ethnic backgrounds are increasingly likely to be in the labor force and to work for more of their adult lives; more than half of all women are now in the labor force in any year. The vast majority of these women work in predom- inantly female occupations. The most seri- ous consequence of this segregation is the persistent wage gap between the sexes. In 1981 white women who were employed fi~-time year-round earned less than 60 per- cent of what white males earned; black women earned 76 percent of what black men earned; and Hispanic women earned 73 per- cent of what Hispanic men earned. In ad- dition, black women earned 54 percent and Hispanic women earned 52 percent of what white men earned. Approximately 35-40 percent of this wage gap can be attributed to occupational segregation, and sex segre- gation within occupations apparently ac- counts for much of the remaining disparity. Hence, in the absence of occupational-ieve! segregation, women would earn about 75 cents for every dollar a man earns rather than the well known 59 cents, and in the absence of job-level segregation the gap would be still smaller.) These economic consequences must be considered in light of the fact that the ma- jority of employed women either support themselves and their clependents or con- tribute their earnings to the income of fam- ilies in which husbands have below-average earnings. This is especially true for minority women. Working in a preclominantly female occupation lowers the wages of both female and male workers, but women in such oc- cupations, on average, earn even less than their male counterparts. The consequences of women's income loss due to job segre- gation extend beyond their years in the labor force. Because segregation depresses their wages, upon retirement women receive lower social security and pension benefits. Nonwage consequences of sex segregation have been less fully documented, but evi- dence clearly shows that female-dominated occupations provide less on-thejob training ~ This computation is based on data in "20 Facts on Women Workers" (U. S. Department of Labor, Wom- en's Bureau, 1982). 123
124 WOMEN'S WORK MEN'S WORK and fewer opportunities for advancement. Some people have also argued that women's resulting lower income and occupational sta- tus reduce their bargaining power in the household and contribute to the unequal di- vision of domestic work. In addition to these adverse consequences for individuals and their families, human resources are wasted when workers are allocated to jobs on the basis of gender (and often race or ethnicity) rather than ability. Whenever sex segregation reflects bar- riers and constraints rather than choice, it is at odds with American values regarding equality of opportunity. We do not believe that full equality of opportunity necessarily requires the policy goal of eliminating all segregation in employment, so that all jobs would have equal proportions of women and men. How much job segregation would exist in the absence of artificial barriers of various kinds is unknowable, but it is our judgment that job segregation by sex would be sub- stantially reduced if barriers were removed. In this chapter we summarize our find- ings, which appear in somewhat greater de- tai} at the end of each of the preceding chapters, and make recommendations for reducing sex segregation. We also identify issues that require further research. SUMMARY OF FINDINGS Measuring Sex Segregation During the past decade women's occu- pational options have unquestionably ex- panded. Their participation has increased sharply in several occupations previously predominantly male by tradition or policy for example, lawyers, bank managers, in- surance adjusters, postal clerks, bus drivers, and janitors, among others. In other occu- pations, women's representation is small but increasing rapidly, for example: coal miners, police officers, and engineers. The overall index of occupational sex segregation de clined by nearly 10 percent between 1972 and 1981, more than it had during any pre- vious decade in this century. Much of this decline was due to women's increased par- ticipation in many occupations that were 20- 60 percent female in 1970 as well as to the decline in the size of some female-domi- Dated occupations, rather than to the entry of women (or men) into the most atypical jobs for their sex. Nevertheless, sex segregation continues to characterize the American workplace, de- spite the changes that have occurred in some occupations. Millions of women continue to work in a small number of almost totally female clerical and service occupations, and men continue to make up the majority of workers in the majority of occupations. The segregation index computed across several hundred detailed census occupational cat- egories stood at 62 in 1981, indicating that 62 percent of all female or male workers would have to shift to an occupation cur- rently dominated by members of the other sex in order for the distributions of female and male workers across occupations to be identical and for occupational sex segrega- tion to be totally eliminated. This measure of segregation understates the amount of sex segregation in jobs, since it does not capture the considerable amount of segregation within many occupations and across estab- lishments. As successive generations of young black women have found jobs in clerical and professional occupations, rather than in the low-paid service occupations to which their mothers and grandmothers were largely re- stricted, the occupational distribution of black women increasingly resembles that of white women. Racial segregation in the workplace among both sexes has declined sharply over the past 40 years. Younger co- horts of all races experience somewhat less sex segregation in the workplace than the general population. Moreover, over the past 10 years the sexes have increasingly selected similar college majors, and the proportions
FINDINGS AND RECOMMENDATIONS of students in professional schools who are female have increased dramatically. It is too soon to know how these young women en- tering new occupations will fare, but wom- en's very presence in these formerly male occupations signals an important social change that should make the path for their successors easier. At the same time, over the remainder of this decade the overall level of segregation is projected to decline only very slightly- between 1 and 5 points, to somewhat less than 60 in l99~primarily because most of the occupations in which the greatest ab- solute growth in jobs is expected are still predominantly of one sex. Although the pro- jections of employment growth may be in- correct and some sex-neutral occupations may grow especially rapidly, only under the most optimistic assumptions is the index ex- pected to continue to decline at the same rate as it did in the 1970s. The overall per- sistence in the aggregate level of segrega- tion, contrasted with changes that have occurred in some occupations, suggests the metaphor of the half-empty/half-full glass: focusing on the increases in women's oc- cupational opportunities suggests important social changes in women's aspirations and expectations and employers' receptiveness to them; focusing on the continued high lev- els of sex segregation points to the need for greater change in workplaces and other in- stitutions and in the attitudes of employers and coworkers to eliminate the barriers and restrictions that we believe cause much of the sex segregation observed. Explaining Sex Segregation Several explanations have been proposed to account for the persistence of sex segre- gation in the workplace; they emphasize dif- ferent factors and differ strongly in the interventions they imply. Not surprisingly, the evidence neither provides full confir- mation nor warrants full rejection of any sin- gle explanation. The scientific evidence we 726; reviewed, however, fails to support the ar- gument that women's occupational out- comes result primarily from free choices that they make in an open market. It suggests rather that women face discrimination and institutional barriers in their education, training, and employment. Often the op- portunities that women encounter in the la- bor market and in premarket training and education constrain their choices to a narrow set of alternatives. In reviewing explanations for sex segre- gation, we considered the role of cultural beliefs; barriers to employment, including discrimination; socialization, education, and training; family responsibilities; and the op- portunity structure. Cultural Beliefs Beliefs about differences between the sexes that are grounded in Western cultural values contribute to the persistence of sex segregation. These beliefs take as axiomatic that women's primary sphere is the home- and that of men is the workplace-and as- sume innate sex differences in personality and physical characteristics that are sup- posed to suit women and men to different kinds of work and militate against their working together except uncler certain con- ditions. As a result, employers and job seek- ers share attitudes about what kinds of work are appropriate for each sex, and many oc- cupations have come to be labeled male or female. Although the sex labels of particular occupations have changed in the past, the rationalizations for these shifts seldom chal- lenge the underlying assumptions that have resulted in the classification of most occu- pations as either women's or men's work. Yet attitudes about women's roles, their right to do wage work, and appropriate re- lations between the sexes have changed sub- stantially over the past 40 years, hand in hand with increased public awareness of changes in women's actual labor market be- havior. Ike growing participation of women
126 WOMEN'S WORK MEN,S WORK in the labor market, particularly large in- creases among women with young children, and the entry of women into highly segre- gated occupations, even in small numbers, challenge implicit assumptions about wom- en's work lives and about appropriate jobs tor women and men. Even small changes signal to future workers that society now permits them to pursue occupations custom- arily held by the other sex in the past. Al- though definitive demonstration is not possible, theories of social change suggest that the increasing recognition by lawmak- ers and the courts of women's right to equal opportunity in the workplace reinforce at- titudinal changes. As a result, as integration takes hold, occupational sex stereotyping declines, and policies that facilitate the movement of persons of either sex into sex- atypical occupations should foster farther declines in segregation. r Barriers Despite recent changes in attitudes and new challenges to old beliefs, we found that a variety of barriers legal, institutional, and informal still limit women's access to oc- cupations in which men have customarily predominated. These include recruiting sys- tems that either depend on worker referrals or hire from male-dominated preemploy- ment settings (e.g., vocational education classes, the military); requirements for non- essential training or credentials that women often lack; veterans' preference policies; promotion and transfer rules, such as de- partment- rather than plantwide seniority systems, that hamper women's movement between jobs and departments; preemploy- ment barriers to relevant job training, such as age restrictions for apprenticeship; and factors such as work climate, harassment, and sponsorship. Employers' acceptance of cultural stereotypes about the appropriate gender for certain jobs or their beliefs about women's and men's characteristics lead some to discriminate to consider gender in hir- ing workers and assigning them to jobs. Sta tistical discrimination Me practice by which employers judge the costs of potential em- ployees based on beliefs about the groups to which they belong may also play an im- portant role in narrowing women's occu- pational opportunities, although its effect has not been quantified. The weight of scientific evidence indicates that discrimination has played a significant role in maintaining a sex-segregated work force. That women believe they face dis- crimination is evidenced by the tens ofthou- sands of sex discrimination complaints filed under Title VII of the 1964 Civil Rights Act (which prohibits sex discrimination in many employment practices). A number of statis- tical studies of large employers show that equally qualified men and women are often assignee] different jobs, with long-term ef- fects on their subsequent careers. Case stud- ies of some employers against which com- plaints have been filed and of certain in- dustries provide corroborative evidence of the occurrence of sex discrimination in em- ployment practices. Socialization, Education, and Training On the supply side, socialization, edu- cation, and training are important because they affect the attributes and qualities that people bring to the labor market. Early so- ciaIization is thought to contribute to sex segregation at work because it appears to lead to the development of sex differences in personality traits and skills that may be relevant for certain occupations; to sex dif- ferences in values, aspirations, and prefer- ences; and to differences between the sexes in knowledge about occupations. But the link between socialization and occupational outcomes and the direction of causation are not well established. For example, consid- erable evidence suggests that perceptions of occupational opportunities influence ex- pressed preferences. We have also seen that sex-typed aspirations and preferences change. Among adults, people pursue sex- atypical occupations when new opportuni
FINDINGS AND RECOMMENDATIONS 127 ties appear. Socialization is a lifelong pro- cess, and adult vocational education, job training programs, apprenticeships, and a return to college or professional schools pre- pare women who previously pursued tra- ditionally female jobs for sex-atypical occupations. The link between education and training and occupational outcomes is better under- stood than is the one for childhood sociali- zation. Premarket training or education is required for many occupations. Girls and boys are highly segregated in most voca- tional education programs, girls take fewer science and mathematics courses than boys in high school, and the sexes, on average, pursue different majors in college. In the past choices to pursue education and train- ing have been subject to considerable con- straint. Parents, teachers, and counselors may treat girls and boys differently and hold different goals for them. Tracking still occurs within the schools, as does sex stereotyping in educational materials. But these differ- ences are declining. High school courses taken and college majors selected have changed for young women and men, and, during the past 10-15 years, admissions pol- icies ant] people's perceptions have changed so that many women now apply to and are accepted at professional schools in numbers almost equal to men. Family Responsibilities Responsibility for the daily care of family members, which women bear more than men, also undoubtedly affects labor market outcomes in many ways, but its link specif- ically to sex-segregated occupations is less clear. One hypothesis, based on human cap- ital theory, is that women choose female- dominated occupations because those oc- cupations are more compatible with child- rearing (by penalizing work interruptions less than male-dominated occupations); this hy- pothesis has found equivocal empirical sup- port. Nevertheless, it seems likely that some people, particularly women, clo seek jobs that they believe are compatible with raising families. Further research is warranted on connections between employment oppor- tunities and family responsibilities for both women and men. Opportunities Sex-role socialization, education, train- ing, and considerations about the compati- bility of various jobs with domestic responsibilities all undoubtedly contribute to the employment decisions that female workers make, but a variety of evidence in- dicates that the occupational outcomes of most workers of both sexes largely reflect what jobs are available to them. Preferences change, stereotyping recedes in impor- tance, and cultural beliefs are transformed as opportunities develop. A striking example is coal mining. When litigation and afErm- ative recruitment efforts opened jobs to women, large numbers of women sought work as miners. We believe the occupational opportunity structure that the labor market presents to workers is key to understanding the perpetuation of sex segregation. The col- lective, cumulative actions of employers cre- ate an opportunity structure that strongly influences workers' preferences, knowI- edge, and occupational outcomes. Employ- ers also respond to changes in workers' behavior and alter their policies accordingly. Of course, all of the factors we have dis- cussed are interrelated. As beliefs have changed about the jobs that women might hold, young women's occupational aspira- tions have become less sex-typed. That their behavior follows suit is seen in the unprec- edented numbers of women training for what were formerly almost wholly male profes- sional occupations. Employers, too, have re- spondec! to women's changing attitudes and behaviors. Because broader opportunities and diminished barriers have been accom- panied by changing cultural values and heightened consciousness regarding gender equity, it is difficult to judge their relative impact. Nevertheless, we place central im
128 portance on removing the remaining bar- riers that prevent women from exercising free occupational choice and enjoying equal employment opportunity in the labor mar- ket. Reducing Sex Segregation Laws and regulations instituted in the 1960s and 1970s prohibit sex discrimination in employment and apprenticeship pro- grams and mandate sex equity in federally funded job training programs and vocational and general education. Although their im- plementation was often slow and enforce- ment sometimes inadequate, the evidence inclicates that when leadership has been strong, employers and educators have had adequate incentives and resources have been allocated to eliminate barriers, women have made substantial progress in entering some predominantly male occupations and training and educational programs. Definitively establishing that women's gains were caused directly by interventions is quite difficult, however. On one hand, the very existence of antidiscrimination laws or regulations may contribute to change. Ac- cording to one theory underlying law en- forcement, most change occurs through voluntary compliance by establishments against which no action has been taken, either out of the desire to avoid sanctions or be- cause laws help to reshape employers' opin- ions about acceptable behavior. At the same time, laws encourage women to believe that they will not face discrimination and hence to train for and pursue sex-atypical occu- pations. On the other hand, important changes-including women's heightened consciousness of their rights and possibili- ties, prompted by the feminist movement occurred during the period in which most interventions were implemented and were an important force for their enactment. Ob- viously, disentangling such cultural changes is difficult. Some of the studies that attempt to demonstrate the impact of specific laws or regulations are imperfect. Taken to WOMEN'S WORK, MEN'S WORK "ether, however, the case studies and sta- tistical research present a compelling case for the long-term effectiveness of legislative remedies. The decreases in federal enforcement ef- fort that have occurred since 1981 and recent changes in the philosophy of enforcement including reversals offederal civil rights pol- icy in some areas, are likely to affect wom- en's future employment opportunities. It remains to be seen what the effects will be. The evidence in this report suggests that the remedies introduced in the preceding two decades have generally reduced segrega- tion. The committee is concerned that de- creased federal effort and changes in policy will have negative effects, particularly be- cause perceptions of reduced effort are likely to affect voluntary efforts. - 7 Interventions in the Workplace Most interventions to reduce segregation that have been implementer] over the past two decades have been directed at the work- place. Data that assess their impact are lim- ited, since evaluation was not a high priority for enforcement agencies. Most accounts of the enforcement process agree that the reg- ulations have a history of uneven and often limited enforcement. Nevertheless, the evi- dence from case studies and statistical anal- yses supports the conclusion that when commitment to enforcement was vigorous and resources adequate, interventions con- tributed to increasing women's access to oc- cupations, industries, and jobs that men have dominated. In particular, women's partici- pation increased in three industries targeted for special emphasis by the Office for Fed- eral Contract Compliance (banking, insur- ance, and mining); contractors in a special program in another industry, construction, admit that increases would not have oc- curred without goals and timetables. Fully implemented goals and timetables also fos- tered women's participation in apprentice- ship programs. Large increases in women's participation
FINDINGS AND RECOMMENDATIONS 129 in several predominantly male professions can be attributed in part to affirmative action programs. Some professional training pro- grams actively recruited women students, and women were more willing to invest in extensive training with the assurance that jobs in the profession would be open to them. Studies of firms against whom suits were filed show increases in the number of work- ers in sex-atypical occupations. This does not mean that segregation has been eliminated in these firms or others covered by statutes and regulations, but progress is undeniable. At the same time, however, many occupa- tions and establishments appear to be almost untouched by the law. Surveys of large es- tablishments indicate that the awareness of federal enforcement by top management has been an important factor in expanding the opportunities available to female employ- ees. Restructuring personnel practices was often necessary to ensure women's access to some jobs. The association between mana- gerial commitment and awareness of the en- forcement of federal regulations supports the importance of strong federal enforcement of antidiscrimination laws. Ibe interventions that were least effective lacked either incentives for compliance or the support of those charged with their im- plementation. The link between these two factors indicates the necessity of including incentives in regulations: sanctions against violators or rewards for those in compliance. Evidence from the construction industry and certain apprenticeship programs demon- strates that goals and timetables are impor- tant for women's entry into male-dominated occupations. Goals create a demand for women that in turn generates a supply of applicants seeking training and jobs. Interventions in Job Training lob training programs also have potential for reducing sex segregation in work. The Comprehensive Employment and Training Act (C ETA) was enacted in 1973 to provide job training and public service employment for the economically disadvantaged, un- employed, and underemployed with a goal of maximizing their employment options. In 1978 an amendment to CETA stipulated that all programs must contribute to the elimi- nation of sex stereotyping. Evaluations of CETA's effects on women prior to the 1978 amendment suggest that it did not facilitate desegregation. CETA did support a few very effective small programs designed to in- crease women's participation in nontradi- tional occupations. These programs demon- strate the potential of federally sponsored training programs to integrate male craft and technical jobs and indicate that such pro- grams can be effective intervention strate gies. Interventions in Education It is more difficult to assess the impact on employment segregation of federal laws passed during the 1970s to eliminate dis- crimination and promote sex equity in gen- eral education. Foremost among Rem was Title IX of the 1972 Educational Amend- ments. To the extent that the educational system contributes to sex segregation in the worlq?lace, legislation aimed at achieving sex equity in education may ultimately promote women's integration into customarily mate jobs. Although women's participation in sev- eral male fields of study during the past clec- ade has increased substantially, no evidence exists that allows us to attribute these in- creases directly to the enforcement of Title IX. Evidence regarding the effectiveness of the 1976 amendments to the Vocational Ed- ucation Act is clearer. These amendments called for vocational education programs that received federal funds to eliminate sex bias and stereotyping in vocational education programs. The effectiveness of this amend- ment has vaned widely by state, suggesting that state laws or federal monitoring are nec- essary for the federal law to reduce sex bias. Where strong commitment was present at the state level, female students' represen
130 tation in mixed or traditionally male pro- grams increased substantially. State pro- grams that were most successful established a broad support base for these women stu- dents and fostered contacts with prospective employers, much like the successful pro- grams Ended under CETA or sponsored by community groups. In both vocational ed- ucation and job training, large-scale inter- ventions have been less likely to show measurable success than smaller, locally run programs, probably because of the difficulty of ensuring replication in implementation. Conclusion Because the causes of segregation-cul- tural values, socialization, sex bias and track- ing in the educational system and in job training programs, discrimination, and in- stitutionalized and informal barriers in the workplace-interact with each other and op- erate together to restrict access to educa- tion, training, and employment in sex- atypical occupations, remedies are most likely to be effective when they address mul- tiple causes (e.g., training combined with placement programs). Chat social values have changed during the same decades as rem- edies have been established is, of course, no coincidence. Changes in values and peo- ple's increasing willingness to entertain non- stereotyped possibilities create a context in which implementing remedies can be par- ticularly elective. POLICY RECOMMENDATIONS Our recommendations are neither new nor startling. They do not detail new programs. Although there is still considerable debate within the scientific community about the causes of sex segregation in employment, there is little disagreement that barriers and constraints play a significant role. Our re- view of the available evidence regarding the effectiveness of the various remedies that WOMEN'S WORK, MEN'S WORK have been used to address sex segregation convinces us that many work when imple- mented properly. Consequently, our rec- ommendations concern improving the enforcement of equal opportunity laws and expanding voluntary efforts in employment and education. The evidence also clearly in- dicates that leadership is a critical compo- nent of efforts to bring about change in this area. Committed leadership, in the execu- tive and legislative branches, in businesses and workplaces, in schools, and in the na- tional media, all contribute to a climate that encourages voluntary change and enhances the credibility of enforcement efforts. Sex segregation is a deeply rooted social and cultural phenomenon. It is perpetuated not only by barriers and constraints, but also by habit and perceptions. Everyone's atti- tudes and behaviors, including women's, are affected by existing occupational patterns; even if there were no obvious discriminatory practices, these patterns would tend to be perpetuated. Consequently, our recom- mendations are addressed to employers and educational institutions as well as to enforce- ment agencies, and they concern family re- sponsibilities as well as employment conditions. Recommendations for Enforcement Agencies Regarding Employment During the past decade sex segregation has broken down in many occupations, and substantial change has occurred in many others. Those changes would have been far less likely without governmental enforce- ment and private litigation. In view of these considerations, any reductions in personnel and budget for important federal enforce- ment agencies may have a negative effect on women's employment opportunities. Strong; enforcement of antidiscrimination laws in employment has been effective in reducing sex segregation in the workplace. If this goal is to be pursued, enforcement agencies such as the Equal Employment
FINDINGS AND RECOMMENDATIONS . 131 Opportunity Commission and the Office of Federal Contract Compliance Programs should continue to play an important role, by sustaining and improving efforts that have been shown to work and by developing new effective approaches. These agencies re- quire sufficient resources to carry out their mandate. Compliance and Enforcement The rate at which sex segregation will con- tinue to decline will be influenced by the magnitude of the efforts and the determi- nation of agencies charged with enforcing antidiscrimination laws and regulations. En- forcement agencies particularly need clear policy direction and committed leadership. Laws and regulations prohibiting cliscrimi- nation must be vigorously enforced to elim- inate remaining barriers to equal employ- ment opportunity in the workplace. Although the selection of particular strategies wid con- tinue to depend on the outcomes of political and judicial processes, goals and timetables have been effective, especially in highly sex- segregated industries, and the committee believes the use of these important tools should continue. Identification of particular areas has been effective and should be con- tinucct, with specific occupations and in- dustries selected on the basis of women's representation in them. While individual claims must be processed efficiently, en- forcement agencies should also pursue larger, more visible cases, because of their impact on other employers, employees, and the public generally. Employers with whom conciliation agreements or consent decrees have been reached should be monitored and adequate fimding should be provided for this purpose. Continued progress in reducing sex seg- regation will depend on the development and implementation of new and more effec- tive enforcement strategies, improved co- ordination between programs focused on different aspects of the problem, and the identification of new approaches to pay eq- uity. A major objection to occupational seg- regation is that to the extent it results from discrimination, it produces wage inequity. Even if strategies to reduce segregation are fully implemented, because of the stability built into the occupational structure, sex segregation wiD continue for a very long time. For this reason, we recommend the explo- ration of efforts to redress wage inequity as well as to reduce segregation. We urge the Equal Employment Opportunity Commis- sion and the Office of Federal Contract Compliance Programs to explore and de- velop enforcement strategies that would en- sure equitable pay for female-clominated jobs whose wage rates have been depressed by discrimination. Although we believe wage equity strate- gies should be explored, we stress that the traditional equal access and affirmative ac- tion approaches are also necessary to ensure equal opportunity in the labor market. Ibe two strategies are related: higher wages for ~ . women s lo as may encourage more men to exacter them, thus enhancing integration, and strategies focused on integration may con- tribute to wage equity. Research and Evaluation Lee enforcement agencies should develop much stronger programs of policy-relevant research on such issues as the sources of change in occupations and industries in which the most rapid change has occurred; newly identified problems, such as sexual harass- ment; techniques for evaluating jobs and other issues related to pay equity; and the extent and causes of resegregation of jobs. In addition, policy evaluation units of en- forcement agencies should study ways to im- prove the effectiveness of enforcement activities, including identifying the features of the most successful compliance or vol- untary activities carried out by employers or others, analyzing the applicability and transferability of successful programs and the
132 need for changes in laws and regulations, evaluating and exploring alternative strate- gies to assist in the most effective targeting of enforcement resources, and developing enforcement and training programs for newly . . emerging issues. Recommendations for Employers Voluntary action by employers has con- tributed a great deal to desegregating cer- tain occupations and workplaces. The most important factors in employer action have been found to be, first, top managerial com- mitment and, second, communicating that commitment to employees at all levels. To this end, incorporating the goal of nondis- crimination in statements of corporate mis- sions and building equal opportunity into a business's public image are vitally impor- tant. Compliance with anti`discrimination laws ant! regulations is essential, but employers can clo considerably more to achieve greater sex integration. Employers can pursue a more systematic approach that includes in- fo`~ing recruiters, managers, supervisors, and employees of the detrimental effects of sex-typed jobs; setting targets; identifying potentially discriminatory mechanisms; es- tablishing monitoring procedures; including responsibility for achieving equal employ- ment objectives in job descriptions; and evaluating and rewarding line managers for fulfilling those objectives. Women appli- cants for and employees in stereotvoicallv female positions should be informed about and offered training for opportunities in pre- dominantly male jobs. At the same time, job requirements and policies regarding pro- motion and on-thejob training (including the organization of seniority systems and col- lective bargaining agreements) should be re- viewed to ensure access. Access and per- formance are also facilitated by the availability of equipment that is comfortable for both women and men and facilities such WOMEN'S WORK, MEN'S WORK as locker rooms and rest rooms that are con- veniently located. Employers should also evaluate job-re- lated compensation criteria and current job evaluation systems for sex bias and revise them if necessary. Compensation equity should extend to benefits, such as pension accruals and parental leave. Employers should also pay close attention to the environment in their workplaces, be- cause it is critically important. Employers can support informal networks that contrib- ute to career advancement and information sharing and attempt to ensure that infor- mation systems are open. Women in atypical positions serve as positive role models, and both formal and informal networks enhance their contribution in encouraging others to follow them. The replacement of sex-biasecl training materials or offensive decorations can also contribute to an improved work cTi- mate. Employers should take appropriate disciplinary action against supervisors and employees who practice or condone sexual or other forms of harassment against women. The business environment can also be af- fected by considerations beyond the work- place. Work-related meetings or company- supported social events that take place at discriminatory clubs can convey a lack of commitment to equal employment oppor- tunity goals and should be avoided. In addition, we recommend that employ- ers explore ways to make work schedules more flexible, with attention to both the . . , workday and the sequencing of career stages, in order to ensure that employees with fam- fly responsibilities have equal access to all occupations and promotion opportunities. We believe that such options should be pro- vided for both men and women across all the occupations and sectors of the firm, and we urge employers to encourage their use by parents of both sexes. To ensure that such policies are effective, organizations should take special care not to penalize men and women who elect part-time, flexible time, or parental leave options.
FINDINGS AND RECOMMENDATIONS 133 We recognize that, as useful as these rem- eclies may be in reducing sex segregation, many employers may be unlikely to pursue them because of the costs involved as well as the inertia of established practices, unless they are encouraged to do so by government policy or by public or employee pressure. This is why the social and political climate that national leaders help to create is so crit- ical. National leaders can point out that the social benefits of remedial action are greater than the apparent private costs. A suppor- tive, encouraging public environment en- courages voluntary compliance with the laws as well as other remedial actions. Employers who take such actions often subsequently find that the benefits to the firm, in terms of improved use of human resources, greater job satisfaction, and lower job turnover, out- weigh the costs of change. Recommendations for Education and Employment Training While enforcement of equal opportunity laws in the employment area has been a major catalyst in improving women's posi- tion in the labor market, women's experi- ences prior to employment have important effects as well. Primary among these are ed- ucation and employment training. Educa- tion and educational policy cannot by themselves eliminate job segregation by sex, of course, but educational authorities and schools at all levels can take measures that will promote integration. Sex equity in the public schools is already a matter of federal law under Title IX of the 1972 Educational Amendment and the 1976 Amendments to the Vocational Education Act. Compliance with these laws is an important first step; the responsible federal agencies must in- crease their enforcement and monitoring ef- forts, which have been very small. Elementary schools should provide girls and boys with nonstereotyped information about a broad range of occupations and en- courage them without regard to sex, race, or ethnicity to begin thinking about the wide range of possibilities they might pur- sue. Schools should also eliminate sex- stereotyped teaching materials dealing with family and home responsibilities or personal interests. Secondary educational institu- tions should develop programs that heighten teachers', counselors', and students' aware- ness of the economic consequences of sex segregation. They should encourage female and minority students to consider the con- sequences of alternative curricula and es- pecially encourage their enrollment in mathematics and science courses that will enable them to prepare for and pursue sci- entific, technical, and professional occupa- tions. The expected growth of new computer- related occupations makes it imperative that girls be encouraged to become comfortable with computer learning at both the primary and secondary levels. All students should be encouraged to enroll in courses on a non- segregated basis, and social and academic support shouIcl be provided to students who select classes that are atypical for their sex. Schools should also try to involve parents by informing them of eclucational and em- ployment opportunities for their children. Many of these recommendations apply also to postsecondary eclucational institutions. In the past, college women's choice of majors and their occupational aspirations became more stereotypically female as they ad- vanced in college. Special academic, voca- tional, and personal counseling can help women to realize sex-atypical aspirations and to encourage others to pursue fields in which women are underrepresented. Academic departments in which women are under- represented should determine whether sub- tle biases tend to discourage them and take action to eliminate the biases. In the area of formal vocational education and employment training, states and local school districts need to make much greater efforts to comply with the Vocational Edu- cation Equity Amendment s mandate to eliminate sex stereotyping. Federally as
134 sisted secondary and postsecondary voca- tional education programs should make sex- integrated vocational education an explicit goal, and programs should allocate resources among vocational education areas to maxi- mize students' training in marketable skills. Vocational educators must inform them- seIves about changing patterns offemale em- ployment, including women's growing work- life expectancy, and about wage differences between stereotypically male and female oc- cupations. Special efforts are necessary to make traditionally male curricula hospitable to women and to encourage women's en- rolIment in historically male courses that provide skills and information usefid for many craft and technical jobs. Links with em- ployers in the community that can provide placement opportunities for graduates are particularly important for sex-atypical pro- grams. Continuing education programs offer women a second chance to enhance their employment options. Such programs should be encouraged to inform returning students about the consequences of occupational seg- regation and to provide counseling regard- ing career moves across sex-typed occupations that would lead to higher earn- ings or better prospects for advancement. Federal training and employment pro- grams should be continued, with increased efforts to prepare women and men for sex- atypical employment. Training must be ac- companied by placement facilities. Program evaluation should be based in part on num- bers of persons trained for and placed in jobs that are not typical for their sex. Counseling and support should be continued after job placement because discouragement and ha- rassment are often problems for women workers in sex-atypical jobs. In addition, women who support themselves and de- pendents should be targeted for training in well-paid nontraditional areas. Age restric- tions should be eliminated from apprentice WOAlEN'S WORK MEN'S WORK ship programs. Affirmative steps to recruit women should be employed when veterans' preference policies operate against them. Recommendations Concerning Family Responsibilides An important factor outside the labor mar- ket that affects women's labor market per- formance and opportunities is the way in which family life is typically organized in our society. Although we have not been able to quantify the degree to which the organiza- tion of family life affects sex segregation in the workplace, it is our judgment that both beliefs and practices concerning family care contribute to segregation, and that better and more flexible child care is critical for change. The care of children and family members still appears to be largely women's responsibility, and this responsibility un- doubtedly conflicts with their entrance into and advancement in a number of occupa- tions that routinely require overtime, job- related travel, or inflexible or irregular hours. Moreover, a widespread belief that women rather than men should be primarily re- sponsible for children and family care prob- ably contributes negatively to attitudes toward women workers and their treatment in nearly ~l occupations and work situations. It is critical that assumptions about women's responsibilities for children and families not be used as a basis for discrimination. Because half of all mothers of preschool chilclren and more Man half of those of school- age children are in the labor force, and fath- ers are nearly universally so, opportunities for combining paid work and parenting that reduce the strain on parents and children are needed. Policies to encourage such op- portunities also serve a broader public policy goal of facilitating the best use of available human resources by reducing constraints. Child care of good quality is the most critical service that needs to be provided. While
FINDINGS AND RECOMMENDATIONS 135 most parents wish to play a major role In caring for their children, doing so is not in- compatible with public or private child care assistance. Child care facilities should be sufficiently varied and flexible to accom- modate both the different needs families ex- perience at various stages in the domestic cycle and varying parental preferences. Es- tablishing flexible alternatives may require an expansion of existing federal policies that involve an array of approaches, such as tax incentives to employers who set up work- site day care centers, small business loans for neighborhood centers, public support of such facilities for low-income families, and larger and more flexible child care income tax credits or deductions for parents. Local and state initiatives can include extending the services that schools provide. While more equal sharing by men and women of child care and other home tasks may await change in gender ideology, policy can affect that process. Workplace policies that allow flexible scheduling of work time, part-time employment for both sexes across ad occupations in a firm, and paternity as wed as maternity leave will help to reduce sex segregation. Such inclusive policies will neutralize the inclination to view such ar- rangements as necessary only for women and may thereby encourage men to experiment with alternative allocations of their time and energy between home and the workplace. We also encourage the ongoing reexami- nation of the fecleral income tax and social security systems to attempt to bring about more equitable treatment oftwo-earner cou- ples compared with single-earner families, because it would also facilitate the reallo- cation of pail! and family time between hus- bands ant! wives. Policies that facilitate more equal sharing wiP enhance the ability of both sexes to combine parenting with paid work without undue hardship and will undoubt- edly advance equal employment opportu- nity as well. DATA AND RESEARCH RECOMMENDATIONS Throughout this report we point to the need for farther research, better data, and improved measurement techniques. Here we review and amplify these points. Data Collection Policies for reducing occupational segre- gation require accurate data on the extent, causes, and consequences of the phenom- enon. The most pressing data needs are the collection of establishment-level data and longitudinal data on individuals; the im- provement of data necessary to assess the effectiveness of statutes and regulations pro- hibiting employment discrimination; the improvement of data on the extent of oc- cupational segregation, particularly by eth- nicity; and the collection of observational data concerning the processes of discrimi- nation and job segregation and the responses of individuals and organizations to change. Establishment-Levet Data Although much research concludes that employer personnel policies are an impor- tant cause of occupational sex segregation, only a small portion of this research is based on establishment-level data. Establishment- leve! data can yield considerable additional information over census-based materials. For example, a study by Bielby and Baron (1984), based on a sample of California firms, re- vealed nearly total sex segregation when em- ployer definitions of job categories were used, a degree of segregation much higher than that found in studies based on census job classifications. At the present time, no nationally representative sample of firms that contains information on job segregation or personnel policies or occupations more gen- eraDy exists. Such a data base would not only
136 allow one to determine the generalizability of Bielby and Baron's findings to the econ- omy at large, but would also permit assess- ment of the impact of specific employer personnel policies on segregation. We rec- ognize that a sample of fimns is not easy to develop. Two important sources of difficulty are the lack of a sampling frame for firms in the American economy and the need to se- cure employer cooperation. The benefits of such a national sample for this and other issues suggest the value of developing this data base. Progress in determining the ef- fects of job segregation on the earnings gap between the sexes will depend on devel- oping sources of more detailed data than the census-detailed occupational cIassifica- tion provides. Data on jobs will necessarily be of limited generalizability, but the poten- tial contribution of up-to-date firm- or es- tablishment-level data outweighs this limitation. At the present time, some estab- lishment-leve} job data are available as the result of disclosure during court cases re- sulting from complaints of discrimination. While some such disclosed information is later sealed, some of it remains available. An extremely useful service that could be provided by a research center or a public agency would be the gathering, editing, and documenting ofthis material (which is some- times available on computer tape) to pre- serve confidentiality and make it accessible and usable by researchers. Longitudinal Data on Individuals Much recent job segregation research based on samples of individual workers has emphasized the importance of careers, job ladclers, and job mobility. Several existing nationally representative longitudinal data bases-the University of Michigan Panel Study of Income Dynamics, the National Longitudinal Surveys at Ohio State Univer- sity, the Continuous Work History Survey, and the National Longitudinal Study of the high school class of 1972 - have been in WOMEN'S WORK, MEN'S WORK - valuable. These sample surveys generally contain 10-15 years' information on the same individuals, but, because such a time hori- zon captures only part of a worker's career mobility, these data bases must continue to be funded so that data for completed careers are available for some cohorts of workers. In addition, consideration should be given to developing a longitudinal sample that would include all job changes and full job event histories, not simply records of job status at year end or time of interview, as is the case with currently available samples. Data to Assess the Effectiveness of Laws Assessing the efficacy of statutes and reg- ulations prohibiting employment discrimi- nation is very difficult. In part this is due to the intrinsic difficulty of distinguishing among various causes of change in a complex society. The problem is exacerbated by the lack of adequate data on the activities of enforcement agencies and their results and by an even greater lack of data about indi- vidual litigation. No systematic data regard- ing complaints filed and their disposition are available. When individuals privately pur- sue their cases through the courts, little is known about settlements, in court or out. It may be possible to develop a sampling method for private cases and to compile data more systematically from state and federal enforcement agencies. More detailed data on patterns of job segregation in individual enterprises would facilitate research on changes that occur in these employment pat- terns as a consequence of intervention by enforcement agencies and would greatly im- prove our ability to assess the effectiveness of equal employment opportunity policies. Data on Race and Ethnicity In order to understand the patterns and effects of occupational segregation, data must be tabulated by race and ethnicity as well as sex. Currently, most tabulations are re
FINDINGS AND RECOMMENDATIONS stricted to white-black distinctions. Yet for particular regions of the country the pattern of occupational sex segregation varies by other ethnic and racial groups as well. Al- though, even in the census and the Current Population Survey, problems of sample size sometimes preclude tabulating data for small population groups in small areas, efforts should be made to provide such tabulations whenever possible. Data From Workplace Studies The collection of observational and inter- view data from intensive studies of work- places should be encouraged. We have very few detailed studies of power in workplaces, especially as it pertains to race, sex, and ethnic differences. By what processes do ctis- crimination and job segregation occur? How are they maintained? Similarly, we have vir- tually no studies of the texture of change within organizations in response to equal employment opportunity policies. Who re- sists and how? Who gains? How meaningful and long-lasting are the gains? What is needed to answer this type of question is close observation of workplaces (and such related organizations as unions or families) and the collection of qualitative as well as quantitative data about people's experi- ences, their attitudes toward change, and their adjustment to new social arrange- ments. Measurement The census occupational classifications need considerable scrutiny. Almost all re- search on job segregation uses the census job definitions. Yet some job groups are cov- erec! in much greater detail than others. To the extent that such differences in coverage are correlated with the sex composition of the job categories, a biased picture of seg- regation is obtained. For example, if male- dominated occupations are categorized in more detail than female-dominated occu 137 pations, men's occupationad mobility rela live to women's may very well be exaggerated. If, however, the census aggre- gates several different segregated jobs into a single category, then a census-based seg- regation index will necessarily understate the degree of segregation. In the 1970 census classification, occu- pations that were at least 80 percent female were about three times as large as those at least 80 percent male (Treiman and Hart- mann, 1981~. Whether this reflects the crowding of women into a few jobs, as some have claimed, or finer census distinctions among jobs dominated by men than among those dominated by women remains an open question. This issue could be investigated by comparing job classifications from firms (ofthe kind we call for above) with the census detailed occupational classification. If the average number of job titles mapping onto each occupational category is larger for those census categories in which most incumbents are women than for those in which most incumbents are men, distortion in the cre- ation of occupational categories, rather than crowding, would be suggested. The utility of making comparisons using disaggregated lists of job titles from other sources, such as the titles listed in the Census Classified In- dex of Occupations, the job titles offered by individuals in response to open-ended ques- tions in sample surveys, and job titles used in help-wanted advertisements, should also be explored. Analysis of what jobs entail may also be warranted. It is not impossible, for example, that common usage results in many vastly different jobs being labeled "secre- tary." In such a case, titles that convey finer filnctional distinctions should be developed and adopted. Research is needed to assess several ex- planations that have been proposed to ex- plain sex segregation. In particular, we recommend further study of how occupa
138 WOMEN'S WORK, MEN'S WORK tional aspirations are formed and how they may be linkec! to occupational outcomes. Al- though we know that the sexes cliffer from childhood in their occupational aspirations, our understanding of both the formation of these aspirations and any effect on occupa- tional outcomes is quite limited. Prospective studies that follow individuals from before their entry into the labor force to well into their careers would clarify the ways that as- pirations and prejudices contribute to seg- regation in the workplace. Of particular interest are the effects of adolescents' oc- cupational aspirations, as well as their ex- pectations, on their early and subsequent occupational outcomes. Although no studies have yet demonstrated a definite link be- tween sex-typed preferences and sex-typed occupational outcomes, it is entirely possi- ble that such a link exists. It would be im- portant to know whether having traditional preferences tends to be correlated with being in sex-typical occupations. Studies of both female and male workers that clarify the nature of any relationships between workers' personal characteristics, occupational values, education, and preem- ployment training, on one hand, and various occupational characteristics (e.g., skill re- quirements), on the other, would also reveal the extent of the match between workers' choices and their subsequent occupations. The connections between workers' family responsibilities, their occupational choices, and their job outcomes also require Farther study. To date, the human capital expla- nation for occupational segregation has re- ceived at best mixed empirical support; di- rect investigation of whether women choose various occupations because they think they will accommodate family care responsi- bilities would contribute to a better un- derstanding of this model. Family respon- sibilities may also affect men's work decisions, and their decision-making processes should also be studied. It wool also be important to learn how and to what extent husbands and fathers influence women's decisions about the work they pursue and the time they contribute to family care. Our knowledge of how labor markets op- erate suggests the importance of occupa- tional information for workers' access to jobs. Not enough is known, however, about how people acquire information about occupa- tions, how the sexes differ in their knowI- edge of sex-typical and sex-atypical occupa- tions, or how this knowledge is linked to their occupational outcomes. More gener- ally, research on how both sexes make de- cisions about vocational training and job search strategies would also bear on the hu- man capital explanations for occupational segregation. Additional investigation is also necessary to determine what kinds of barriers exist in labor markets and within establishments to workers' movement into and their retention in sex-atypical jobs and how they might be improved. In many occupations job training represents the primary vehicle through which workers move up job ladders. Very little is known, however, about differences in the sexes' access to on-thejob training, the type of training available to workers in typically female and male occupations, and whether the benefits of training are similar for the sexes. We encourage in-depth stud- ies of large firms to determine what features of firms' internal labor markets including opportunities for on-thejob trainin~re- strict or foster workers' access to sex-atypical jobs. Such in-depth studies, if extended to workers' and employers' behavior through observation and interview, could also in- crease our understanding of various more general phenomena related to discrimina- tion and opportunity, such as group cohe- sion and informal information networks. One way to shed light on some of these issues would be to study belief systems concerning women and work, in order to understand the structure of beliefs that underlie em- ployer decisions, career selection, and co- worker response. Another important aspect of the problem of discrimination within
FINDINGS AND RECOMMENDATIONS 139 workplaces is discrimination in work-related social life, particularly the conduct of busi- ness in single-sex environments, such as clubs and sports. The reasons for the per- petuation of these single-sex organizations and their importance for work life should be Farther explored. Several studies that we reviewed revealer] considerable mobility by workers of both sexes between sex-typical and sex-atypical occupations. The movement of workers out of occupations not customarily held by per- sons of their sex is consistent with the ex- istence of structural and informal barriers to workers' success and retention in sex-atyp- ical jobs. We recommend systematic case studies that compare the experiences of workers in sex-typical ant! sex-atypical jobs on several dimensions to determine the fac- tors associated with retention in sex-atypical occupations. In general, during the 1970s, women's movement into occupations that men have dominated was not matcher! by men's movement into traditionally female nonprofessional occupations. Additional re- search is needed to determine whether em- ployers resist employing men in these jobs or whether men avoid them because they lack the necessary training, seek higher wages, or fear that they will be stigmatized. It will also be interesting to see whether young women who have entered formerly male-dominated occupations have careers and age-earnings profiles similar to their male counterparts, or whether, like earlier female cohorts, their earnings and career prospects tend to fall off over time. Some observers have speculated that oc- cupational integration may be a temporary phenomenon; they suggest that the entry of substantial numbers of sex-atypical workers into an occupation may ultimately be fol- lowed by its becoming dominated by mem- bers of that sex. As evidence they typically offer historical examples; we found no sys- tematic research on resegregation. Not enough time has elapsed since women's re- cent movement into some formerly male dominated occupations to determine if those occupations will become predominantly fe- male. It is important to learn whether in- tegrating occupations tend to be unstable and if integration or resegregation is accom- panied by change in real wages or occupa- tional status or the development of new barriers to free occupational choices. If re- segregation occurs, it may be because op- portunities remain limited in yet other sectors of the labor market, so that women flock to the few newly available jobs. Whether resegregation is a "second gener- ation" response to efforts at integration war- rants investigation. If women's gains are subverted anew by new mechanisms, the prospects for lasting improvements require new unclerstanding, regardless of the effec- tiveness of current strategies for desegre- gation. Further research is also necessary to de- termine the conditions under which various interventions contribute to reduced sex seg- regation that is sustained in the long run. In particular, we recommend closer study of the channels through which the statisti- caDy observed effects offederal enforcement agencies, such as the Equal Employment Opportunity Commission and the Office of Federal Contract Compliance Programs, oc- cur. Much of the evidence regarding the effectiveness of federal regulations against employment discrimination is based on comparisons of black and white men. Re- search that compares race and sex groups is needed to test whether race effects also hoIcl for gender. We noted above the enormous need for policy-relevant research on such issues as the effectiveness of targeting and the replicability of remedies. We do not know how the intensity of effort is linked to the magnitude of change and which strategies are most cost-effective. Without such knowI- edge we cannot know whether greater en- forcement efforts or different enforcement tools are necessary to further reduce sex seg- regation.