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.
Prospects for Pay Equity in a Changing Economy Pamela Stone Cain INTRODUCTION The increase in women's labor force participation, which has been espe- cially rapid since World WarII, has been characterized as a "subtle revolu- tion" (R.E. Smith, 19791. Equally subtle and dramatic have been other trends, which, together with the increase in women's employment, have resulted in a reconfiguration of the U.S. economy and work force. One source of these trends is the fundamental transformation that has occurred as the United States moves from a manufacturing to an increasingly service- based, more technologically sophisticated economy. A second source of change is postwar demographic trends in fertility, household and family structure, and immigration. In this paper I focus on features of this ongoing transformation as they have particular relevance for the status of women and their chances for achieving parity with men in wages and jobs. To date, women's progress in the labor market has been slow. Despite their increased labor force participation and the passage of legislation to prohibit sex discrimination in employment (the Equal Pay Act in 1963 and Title VII of the Civil Rights Act in 1964), the earnings gap between full- time, year-round male and female workers has decreased only a few percent- age points in recent decades (Norwood, 19841. Moreover, although the job titles held by women have proliferated (U.S. Department of Labor, 1982), they remain concentrated in fewer jobs than men. Will future trends in the labor force and economy tend to accelerate the progress toward parity that 137
138 CAIN women workers have made facilitating job integration and closing the wage gap or will they tend to halt or reverse it? There are no easy answers to this question. Given the scope and complexity of the factors involved, the sometimes contradictory research findings, and the difficulty of predicting the future, different interpretations and outlooks are possible. In this paper I develop alternative scenarios, beginning by reviewing recent and projected trends in women's labor force participation as well as other changes that affect the composition of the work force. Next I turn to a consideration of changes in the U.S. industrial and occupational structure as these affect the opportunities open to women. Finally, against the background of changes in labor supply and demand, I review trends in job segregation and the wage gap to assess women's progress to date and speculate about their prospects for the future. The competing interpretations that I offer highlight what is perhaps the only conclusion to be drawn: that the slow and incremental gains women workers have made cannot be taken for granted, nor can their pro- gress be assumed to proceed as an inevitable corollary of their continued, significant presence in the labor force. I conclude by offering suggestions for needed research that would enable us to better address and illuminate these critical issues. A NOTE ON THE PROJECTIONS USED IN THE PAPER This paper makes use of the economic projections to 1995 compiled by the U.S. Bureau of Labor Statistics (BLS) as the basis for future scenarios (Bureau of Labor Statistics, 1984a). These projections have been criticized for failing to take into account factors in labor force growth such as the attractiveness of jobs as well as for ignoring the complementarily of men's and women's labor supply decisions (Lloyd and Niemi, 1979~. Notwith- standing the validity of these criticisms, BLS projections are nonetheless useful because of their comprehensiveness and accessibility; moreover, they are periodically updated. The most recent projections replace those issued in 1980 and incorporate new information from the last census. Specifically, the 1984 revisions reflect new assumptions about longer life expectancy and higher levels of net migration as well as lower assumptions about fertility levels. These changes result in an assumption of higher overall population growth. With regard to labor force projections, the most notable change in the 1984 estimates occurred for women ages 25 to 34. Although BLS has con- sistently underestimated the participation of this group, in the new projec- tions the labor force participation rate was lowered 2 percentage points to
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY 139 81.7.1 This group still shows the largest projected increase of any labor group. Projected rates for men ages 35 to 54 and women age 35 and older were revised upward (Fullerton and Tschetter, 1984:7-~. The basic methodology used in these forecasts is discussed fully else- where (Bureau of Labor Statistics, 1982b). Briefly, BLS develops three sets of projections, labeled "low," "moderate," and "high" growth. Each set entails slightly different assumptions about fiscal and monetary policy and future growth rates in the gross national product (GNP), given industries, occupations, and segments of the labor force. Labor force projections are based on population projections prepay by the Bureau of the Census. Economic projections are based on an econometric model developed by Chase Econometrics (see Andreassenet al., 1984, for details). In comparing projections generated by alternative models, BLS concluded that their esti- mates of future labor force growth and composition were more sensitive to demographic than to economic assumptions. BLS cautions that "none of the three projections should be favored as the most likely" (Andreassen et al., 1984:9~. Rather, the set of forecasts is intended to generate a reasonable range of possibilities. For purposes of this paper and following BLS convention, I present the estimates from the mod- erate growth scenario. These assume continued economic recovery from the recent recession, with real GNP increasing at an annual average of 3.2 percept through 1990, falling to 2.5 percept through the mid-199Os. For the two periods, unemployment rates of 6.3 percent and 6 percent are assumed, respectively, which seem low in light of recent experience. Given the past 1 A set of projections of women's labor force participation developed by Waite (1978), which takes into account the influence of changing sex-role attitudes as well as demographic consider- ations, predicts an increase in labor force participation rates from 1975 to 1990 in the range of 0.7 to 5.7. From 1985 to 1990, BLS's moderate growth projection shows a change of 5.7 points. Projec- tions by R.E. Smith (1979), which take account of women's marital status and childbearing as well as other demographic factors, are compared below with the BLS projections used in this paper: Female Labor Force Participation Rate Projected to 1990 Smith BLS Ages 1~24 67.1 69.1 Ages 25-54 68.6 75.6 Age 55 + 21.2 20.5 Total 54.8 58.3 Smith's projections also agree substantially with those of BLS, which are slightly higher, espe- cially among women ages 25 to 54. These comparisons indicate that in its most recent set of projections, BLS may have taken steps to remedy its perennial underestimation of women's labor force participation.
140 CAIN record of BLS of underestimating women's economic activity and because there is error inherent in any such projection, the data presented below should be interpreted broadly, as indicative of general trends and patterns rather than as precise estimates of absolute levels and rates. THE CHANGING DEMOGRAPHY OF THE LABOR FORCE Four major factors shape the composition of the future labor force. First, historical shifts in fertility have resulted in a cycle of baby "boom" and "bust," with corresponding differences in the size of birth cohorts. A second factor is the different age structures and fertility of population subgroups; black and Hispanic populations are younger, exhibit higher fertility, and consequently have higher rates of growth than whites have. A third is the changing roles of women: increasing numbers are entering the paid labor force and becoming heads of households. Finally, recent upswings in the number of immigrants (both documented and undocumented) have an impact, which is difficult to measure, on the age, sex, and ethnic profile of workers. Table 1 shows civilian labor force participation rates of different groups for l9X2 and projected to 1995. Following an already-established trend toward convergence, the labor force participation rate of women is expected to increase from 52.6 percent in 1982 to about 60.3 percent in 1995. Over the same period, men's participation rate is expected to decline from 76.6 per- cent to 76.1 percent. The rate of convergence is most pronounced among prime-age workers ages 25 to 54. As a result ofthese changes, BLS forecasts that women will account for nearly two-thirds of labor force growth through the mid-199Os, i.e., two out of three first-time labor force entrants or reen- trants will be women. For both blacks and whites, increases in labor force participation will be greater among women than men, although increases for black women, whose labor force participation rates have been high historically, are not as large as those seen for whites. One out of four new or returning workers will be of black or minority background. One implication of increased participation among women ages 24 to 54 is that more and more mothers of young children will be working. Generally there will be increasingly smaller differences in labor force activity among women by marital and family status. From 1950 to 1983, the labor force participation rates of married women with husband present and children under age 6 increased fourfold, from 11.9 to 49.9 percent. Those of women with children ages 6 to 17 more than doubled, from 28.3 to 63.8 percent (Waldman, 19X41. In a special set of projections developed by BLS for women ages 20 to 34 (an age group for which differentials by marital status
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 1 Civilian Labor Force Participation Rates by Sex, Age, and Race, 1982 and Projected to 1995 Percentage Participation Rate Difference Labor Group 1982 1995 1982-1995 Men 76.6 76.1 - 0.5 16-24 72.6 74.5 + 1.9 16-19 56.7 62.9 +6.2 20-24 84.9 84.1 -0.8 25-54 94.0 93.4 -0.6 25-34 94.7 93.1 - 1.6 35~4 95.3 95.3 0.0 45-54 91.2 91.1 -0.1 55 and over 43.8 35.3 - 8.5 55-64 70.2 64.5 -5.7 65+ 17.8 13.3 -4.5 Women 52.6 60.3 +7.7 16-24 62.0 71.6 +9.6 16-19 51.4 58.2 +6.8 20-24 69.8 82.0 + 12.2 25-54 66.3 78.7 + 12.4 25-34 68.0 81.7 +13.7 35-44 68.0 82.8 + 14.8 45-54 61.6 69.5 +7.9 55 + 22.7 19.9 -2.8 55-64 41.8 42.5 +0.7 65 + ~ 7.9 7.0 -0.9 White 64.3 68.1 +3.8 Men 77.4 77.0 -0.4 16-24 74.9 79.1 +4.2 25-54 94.9 94.5 -0.4 55+ 44.2 35.6 -8.6 Women 52.4 60.0 +7.6 16-24 64.7 75.4 + 10.7 25-54 66.1 78.7 + 12.6 55+ 22.4 19.5 -2.9 Black and other 61.6 65.7 +4.1 Men 1.0 70.6 - 0.4 16-24 60.0 52.7 -7.3 25-54 88.0 87.2 -0.8 55 + 40.5 32.6 -7~9 Women 53.9 61.7 +7.8 16-24 48.8 55.3 +6.5 25-54 67.9 78.7 + 10.8 55 + 25.5 22.8 -2.7 Total, age 16+ 64.0 67.8 +3.8 SOURCE: Fullerton and Tschetter (1984:Table 1). 141
142 CAIN have been greatest), married women's labor force participation is expected to increase almost 20 percentage points from 61.6 percent in 1982 to 80.3 percent in 1995. Over the same period, the rate for unmamed (including single, divorced, widowed, and separated) women will increase only 6 percentage points, to 83.2 in 1995 (Fullerton and Tschetter, 19841. Thus, in less than 20 years, mamage and childbearing are expected to exert almost no inhibiting effect on women's participation in the labor force. Other changes in the household and family, especially the increase in single parenting, have implications for women's future labor force activity. In 1983, 59.6 percent of women with families worked, including 55.2 percent of women with children under age 6 (Johnson and Waldman, 1984~. Moreover, from 1970 to 1983, the number of female heads of household in the labor force doubled. According to Johnson and Waldman, the record numbers of marriages, divorces, and subsequent labor force entry during this period are, like other changes in the labor force, a function of the activities of the baby boom cohort. By the 1980s, "divorceeswho have the highest LFP [labor force participation] rate of any marital category had replaced widows (who have the lowest) as the largest group of women maintaining families" (Johnson end Waldman, 1984: 16~. The resulting composition of the labor force is shown in Table 2. With women accounting for nearly two-thirds of labor force growth, they are expected to increase their proportion of the labor force from 43 to 46 percent. Minority representation is expected to grow slightly, to 14 percent. The proportion of white men in the labor force is projected to decline by 4 percent by 1995. By 1983 their majority had already been eclipsed: they made up only 49.8 percent of the work force (Serrin, 1984~. The age structure of the labor force is also changing. As the baby boom cohort matures, prime-age workers will make up a larger share of the work force, and the median age of labor force participants will increase: from 34.8 years old in 1982 to 37.3 in 1995 (Fullerton and Tschetter, 19841. Simulta- neously, younger workers ages 16 to 24 (the baby bust cohort) will decline both in absolute numbers and proportionally. In addition, because of race and ethnic differences in fertility and age structure, blacks and other minori- ties will make up a greater fraction of the future youth work force. Immigration has also contributed to changes in labor force composition. Since the 1930s, immigration has increased fivefold, and in recent years the annual legal inflow has been about 2.2 immigrants per 1,000 population (Chiswick, 1982), an average of 400,000 people per year (Keely, 19791. Not only has the pace of immigration quickened in recent years, but its nature has changed as well. Since the late 1960s, nations in Asia, Latin America, and the Caribbean have become the most important countries of origin, accounting for61 percept ofimmigrants over the period 1971 to 1978
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 2 Labor Force Composition: Percentage Distribution of Different Race, Sex, and Age Groups, 1982 and Projected Labor Group White Men 16-24 25-54 55+ 47 8 34 6 Women 16-24 25-54 55+ Total Black and other Men 16-24 25-54 55+ Women 16-24 25-34 55+ Total 1982 1990 1995 50 10 32 7 46 7 34 6 37 9 23 s (87%) (86%) (86%) 39 7 27 4 7 2 4 6 4 39 7 28 4 7 s s 7 l 6 1 1 1 (13%) (14%) (14%) NOTE: Percentages may not add to 100 due to rounding. SOURCE: Computed from Fullerton and Tschetter(1984:Table 1). 143 (Chiswick, 19821. Moreover, undocumented or illegal immigration has substantially increased. Estimates of illegal immigration vale widely (Keely, 1979), but the Bureau of the Census reports that there are between 3.5 million and 6 million illegal aliens in the United States, about half of them from Mexico (Chiswick, 19821. Thus, recent immigrants are young, of Hispanic or other minority background, and, by virtue of the self-selec- tive nature of migration, have high rates of economic activity. Key features of the labor force of the future can be summarized in light of these trends: · Continued erosion of the prototypical white male worker's share of the labor force; · Maturing (but not graying) of the work force as the baby boom cohort ages and enters the years of prime work life; · A relative shortfall in the supply of younger workers or new labor force entrants;
144 CAIN · A reinforcement of already-established trends in female labor force participation by marital and family status, with more and more wives, moth- ers of young children, and heads of household entering or reentering the labor force; and · An unknown impact of immigration, but one probably adding to the numbers of younger and minority workers. TRENDS IN LABOR FORCE ACTIVITY AMONG EMPLOYED WORKERS Although there are no projections available for indicators of labor force activity among employed workers, a review of recent trends gives some idea of where things might be moving. Hours and Weeks Worked The BLS projections assume a continued increase in part-time employ- ment across all employed workers through the 1990s, with a drop in average weekly hours from 35.1 in 1982 to 33.1 in 1995 (Personick, 19841. Reflect- ing this trend, part-time work (defined as fewer than 35 hours per week) is a small but growing phenomenon among employed men (see Table 31. Corre- spondingly, there has been a drop in the length of men's average work week. Although most employed women, like men, work full time, women have been about two or three times more likely than men to work past time. Women, however, are slightly more likely than men to be working on part- time schedules involuntarily because of slack work, the inability to find a full-time job, or other reasons (U.S. Department of Labor, 19821. TABLE 3 Part-Time Status and Average Hours of Adults at Work in Nonagricultural Industries, by Sex, 1968-1980 Men Women Average Total Average Total Weekly Hours Weekly Hours Year % Part Time at Work % Part Time at Work 1968 5.2 43.5 23.5 35.6 1970 6.3 42.6 24.5 34.9 1972 6.4 43.0 24.7 35.3 1974 6.7 42.6 25.0 35.1 1976 7.3 42.4 25.2 35.0 1978 6.9 42.8 24.7 35.3 1980 8.0 42.1 24.9 35.3 SOURCE: Computed from Bureau of Labor Statistics (1982a:Table B-22).
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 4 Percentage of Part-Time Employment Among Women, by Marital Status, March 1983 Marital Status Never reamed Mamed, husband present Other marital status Mamed, husband absent Widowed Divorced % Part Time 3s.3 29.7 19.5 20.1 33.7 13.3 SOURCE: Computed from Bureau of Labor Statistics (1984b:Table B-1). 145 The increase in women's labor force participation over the last 20 years has not been accompanied by a shift to more full-time work. Over the period 196X to 1980 (see Table 3), the proportion of women on part-time schedules remained steady at about 25 percent, as did their average hours worked per week at about 35. This reflects in good measure the activities of married women. Marital status is, in fact, a majordeterminantofhours worked. In 1983, 83 percent of women who were single heads of household worked full time Johnson and Waldman, 1984~. In contrast, among women with spouse present, only about 70 percent did so. Part-time work was more prevalent among never-married women (on average, these are younger workers, many of whom are also in school), widows, and married women with husband present than it was among divorced women or married women with husband absent (see Table 41. For married women, the presence of children also increases the likelihood of part-time work. In 1980 almost 40 percent of those with children under age 18 worked part time (see Table 51. Although there was little variation in part-time employment by age of children, mothers of preschool-age chil- dren, who are presumably younger then those with order children, were only slightly more likely than mothers of school-age children to work part time, reflecting a general trend toward greater labor force attachment among younger women workers. It should be noted that the large sex differences in part-time and full-time status do not translate into big differences in the number of hours actually worked. As Table 3 shows, in l9X0 women worked an average of only about 7 fewer hours per week than men. Moreover, the foregoing discussion understates somewhat the work time of both women and men, because hours worked are reported for the primely job. In l9X0 roughly 6 percent of employed men and 4 percent of employed women held two or more jobs. Among men the proportion of multiple jobholders has remained fairly stable
146 CAIN TABLE 5 Percentage of Part-Time Employment Among Mamed Women With Work Expenence, Husband Present, 1980 Women with: No children under age 18 Children under age 18 Children ages 6- 17 Children under age 6 Children ages 3-5 Children under age 3 Total % Part Time 26.9 37.4 36.8 38.2 37.6 38.6 32.8 SOURCE: Bureau of Labor Statistics (1983b:Table B-8). TABLE 6 Year-Round Work Experience of Men and Women: Percentage Working 50 to 52 Weeks Part Time or Full Time, 1950-1981 Year Men Women 1950 1955 1960 1965 1970 1975 1980 1981 68.5 71.5 68.4 71.8 70.5 68.2 69.6 69.1 45.0 48.1 46.9 47.9 50.7 53.1 56.6 57.0 SOURCE: Computed from Bureau of Labor Statistics (1982a:Table C-2). since the 1960s; among women it has nearly doubled (Bureau of Labor Statistics, l9X2a: Table C-171. Increasingly, women are seeking additional jobs to supplement part-time work orlow-paying full-time jobs (Sekscenski, 1981), and these extra hours of work are not reflected in the statistics cited above. In contrast to data on hours worked, data on women's year-round attach- ment to the labor force (weeks worked per year) reveal notable changes and a narrowing of sex differences. While the proportion of men working year- round has held steady since 1950 at about 70 percent (see Table 6), the proportion of women doing so has increased substantially from 45 to 57 percent. As with hours worked, the year-round labor force attachment of women is affected by marital status and the presence of children. As Table 7 shows, among full-time workers the year-round attachment of both married and ever-married women has increased substantially since 1960. Although the
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 7 Women With Year-Round Work Experience: Percentage Who Worked Full Time, 50 to 52 Weeks, by Marital Status, 1960-1981 147 Mantal Status 1960 1970 1980 1981 Single 38.6 33.6 37.6 37.3 Marned, spouse present 33.0 40.3 44.1 45.0 Widowed, divorced, separated 46.9 51. 1 55.9 56.0 SOURCE: Computed from Bureau of Labor Statistics (1982a:Table C-3). rate of change among married women was greater than that for widowed, separated, or divorced women, the latter group maintained significantly higher levels of year-round activity throughout the period. Among married women with spouse present who were full-time workers in 1980, 53 percent of those with no children under age 18 worked 50 to 52 weeks, while for those with children the comparable figure was 37 percent. Among women with children under age 6, only 31 percent worked year- round (Bureau of Labor Statistics, 1983b:Table B-81. Combining part-time and full-time workers, a similar pattern is seen: women with children, especially preschoolers, have a tendency to work fewer weeks per year (Bureau of Labor Statistics, 1982a:Table C-171. Other Indicators of Labor Force Attachment Changes in women's labor force participation rates and year-round attach- ment translate into the accumulation of more work experience and longer work lives. Successively younger cohorts of women have higher participa- tion rates. Moreover, new labor force entrants appear to be a declining proportion of all women workers (Lloyd and Niemi, 19791. In 1950 the average work-life expectancy of a 20-year-old woman was 14.5 years. In 1977 a 20-year-old could expect to spend 26 (almost twice as many) years working (S.J. Smith, 19X21. Men's work-life expectancy, meanwhile, dropped from 41.5 years to 37 years. Given sex differences in life expect- ancy, these numbers represent an increase from 27 to 65 percent of women's life-span versus a decrease from 85 to 72 percent of men's (Bureau of Labor Statistics, 1983a: 12~. Thus, by this indicator, too, there is growing similar- ity between the labor force profiles of men and women. 2 2 The methods used to estimate average work life expectancy have undergone revision over the years, so these comparisons are intended to be suggestive only. Recent work-life estimates use an increment-decrement working life table approach that takes into account the age-specific mortality and labor force entry and exit rates that prevailed in 1977 (S.J. Smith, 1983:36). See S.J. Smith (1982, 1983) and Pinch (1983) for a discussion of the relative strengths and weaknesses of current methodology.
148 CAIN Differences between men and women still persist, however, in length of service with a particular employer job tenure. Although sex differences in tenure are narrowing, and they no longer exist among workers up to their mid-30s, as of 1983 average tenure for employed women was 3.3 years compared with 5.1 for men (Mellor, 1984~. A recent analysis of sex differ- ences in turnover found that, while women were more likely than men to quit a job to leave the labor force, their intralabor-force mobility (leaving one job to take another) was actually lower than that of men (Haber et al., l9X3~. Moreover, these results corroborated others (Viscusi, 1980; Blau and Kahn, l9X1) in finding that when differences in type of job, wages, part-time status, age, and family characteristics were taken into account, women did not have higher separation rates (which included permanent layoffs) than those of men. Facilitating this trend toward increased work experience is growing convergence in the unemployment rates of men and women (Bureau of Labor Statistics, 1983a), which has been attributed to women's location in what are sometimes called recession-proof industries. Women are also increasing their education and, in terms of median years of schooling, are now on a par with men. Although they continue to surpass men in rates of high school graduation, they still lag behind them with regard to college and postgraduate training (see Table 8~. In summary, although there is increasing similarity in the labor supply of employed men and women workers, differences persist. This is especially true with regard to part-time work, where marital and family responsibilities still constrain the ability or desire of married women with children to work full time, while impeding husbands and single heads of household to do so. Women are improving their standing relative to men on various productiv- ity-enhancing characteristics such as work experience and education, pro- gress being greatest among younger cohorts of workers. EVOLUTION IN THE ECONOMIC STRUCTURE Industry Outlook Women workers have long been concentrated in particular industries (Norwood, 1984~. For most of these industries BLS projects considerable growth. Senrice industries are expected to create three out of every four new jobs between now and 1995 (Personick, 19841. Within the service sector, miscellaneous services will have the highest growth rate and provide the most new opportunities. This sector includes jobs in business services, such as management consulting, security, and data processing. Finance, insur- ance, real estate, and health and medical services are also expected to show
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 8 Educational Attainment of the Civilian Labor Force, by Sex, Selected Years, 1959-1981 Percentage Distnbution 149 Median Elementary High School College School Fewer than 5-8 1-3 4 years 1-3 4 years Years Year 5 years years years years or more Completed Males 1959 6.1 26.9 20.2 27.2 9.1 10.4 11.5 1965 4.4 21.3 19.4 32.0 10.5 12.4 12.2 1970 2.9 16.9 17.5 35.1 13.5 14.2 12.4 1975 2.2 11.1 17.6 36.2 15.6 17.3 12.5 1980 1.5 8.3 16.0 36.5 17.7 20.0 12.7 1981 1.5 7.9 15.4 37.5 17.4 20.3 12.7 Females 1959 3.5 21.5 19.1 38.1 9.7 8.1 12.2 1965 2.4 16.6 18.7 41.9 10.4 10.0 12.3 1970 1.5 12.2 16.9 45.5 13.2 10.7 12.4 1975 1.0 8.1 17.6 44.7 15.4 13.3 12.5 1980 .7 5.5 14.9 45.0 18.1 15.9 12.6 1981 .8 5.1 14.2 45.5 18.6 15.7 12.7 SOURCE: Bureau of Labor Statistics (1982a:Table C44). above-average growth. Manufacturing industries, most hurt by the recent recession, are expected to rebound and to maintain a steady share of total employment of about 17 percent. Construction-related industries are also expected to rebound, growing slowly at 1 to 2 percent per annum through 1995. Electronic, computer, and other high-tech manufacturing industries are forecast to increase faster than total employment, but the total number of new jobs added to the economy by these industries is relatively small. Estimates of their contribution range from 3 to 17 percent of all new jobs through 1995 (Riche et al., 1983~. Growth, although slowed from previous years, is expected in wholesale and retail trade and in government. In the trade sector, eating and drinking establishments will provide a large number of new jobs. In government, a slowdown in job growth reflects budget reductions of the 1970s and declin- ing school enrollments. As the children of the baby boom cohort enter their school years, an increase in the number ofjobs in teaching is expected. The only industries for which job loss is projected are agriculture and private household services.
150 Recent Trends CAIN Occupational Outlook These broad industry directions, coupled with the introduction of new technologies, have resulted in several occupational trends, many of which gained momentum during the 1970s. In a study of "occupational winners and losers" over this decade, Leon (1982) found that winners occupations that either added many jobs to the economy or that grew by 75 percent or more were primarily in professional, managerial and administrative, tech- nical, and clerical fields. Both criteria (large absolute numbers and high growth) were met by computer-related occupations, health technicians, and bank tellers. Job losses of 60,000 or more workers were posted primarily by blue-collar occupations. Big losses were seen among such occupations as delivery workers, operatives and assemblers in nondurable-consumer- goods-producing industries such as textiles, and in certain sectors of whole- sale and retail trade, e.g., gas station attendants. As a result of these trends, by 1980 while-collar workers made up 50 percept ofthe work force. As Leon (1982:19) notes, women accounted for a disproportionate share of employ- ment growth over the 1972-1980 period: 65 percent as compared with their 38 percent share of total employment in 1982. Many of the occupational winners were female-dominated occupations, and women accounted for virtually all the employment growth in such occupations as secretary, cash- ier, bookkeeper, bank teller, and welfare and health aide. In some cases (e.g., bookkeepers end leaching aides), this growth resulted in an increase in the already high proportion of women in these jobs. Women also accounted for a large share of the increase in employment in male-intensive occupations. Thus, for example, women made up two-thirds of new accountants and 15 percent of new employment among engineers. They also made large relative gains among lawyers, computer specialists, and real estate agents and brokers. Even among slowly growing male- dominated jobs women made inroads. Thus, although only one in five new craft workers were women, their total representation rose from 3 .6 percent in 1972 to 6 percent by 1980. Similarly, small but nonetheless real gains were made in other traditional male occupations, such as truck drivers and warehouse laborers. Projections The current distribution of employment across major occupational groups is compared to the projected distribution in Table 9. For the 1980s and 1990s BLS predicts that professional and technical employment will continue to
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY TABLE 9 Percentage Distribution of Employment, by Major Occupational Group, 1982 and Projected to 1995 Occupational Group 1982 1995 16 Professional, technical and related workers Managers, officials, and proprietors Salesworkers Clerical workers Craft and related workers Operatives Service workers Laborers, except farm Farmers and farm workers 9 7 19 13 16 6 3 17 10 19 12 12 16 6 2 NOTE: Percentages may not add to 100 due to rounding. SOURCE: Silvestn et al. ( 1984:Table 1 ). 151 increase at a rate greater than overall employment, fueled in part by the growth of high-tech industries. The rates of growth of managers, sales workers, and craft workers are expected to keep pace with overall growth. Operators and laborers will continue a long-term decline. As a source of employment displacement and reconfiguration, the impact of new technologies on the job prospects of women (and men) is difficult to assess. Newer technologies appear to be moving with great speed and affect- ing a broad array of occupations (Serrin, 19831. Levitan and Johnson ( 1982) estimate that existing robotics technology could currently perform 7 million jobs, one-third of all manufacturing employment. Although any discussion of this subject is speculative at this point, clearly these technologies have implications for both "women's work" in the office and "men's work" on the shop floor. Already BLS projects that during the l9SOs and 1990s office automation will begin to make significant inroads into clerical employment, with the result that these jobs will grow at slower rates than previously, although keeping pace with total job growth. At the level of detailed occupations, Table 10 shows the 20 occupations that will grow most rapidly through 1995. Dominating the list are jobs in computers, newly emerging industries in data processing and business, and health services. Many of these jobs require a college degree or technical training. Three out of every four jobs for which data are available are dominated by either men or women. Table 11 shows the 20 top jobs in terms of growth in the number of new jobs. The list is led by more traditional occupations that have historically been large employers, for example, clerks, lower-level service workers, retail clerks, and truck drivers. Only a quarter of these jobs require a college degree (Silvestri et al., 1984:42), and four out of five are highly segregated by sex. The 20 jobs listed in Table 10
152 TABLE 10 The 20 Fastest-Growing Occupations, 1982-1995 CAIN Occupation 1. Computer service technicians 2. Legal assistants 3. Computer systems analysts 4. Computer prog~mrners 5. Computer operators 6. Office machine repairers 7. Physical therapy assistants 8. Electrical engineers 9. Civil engineering technicians 10. Peripheral data-processing equipment operators 11. Insurance clerks, medical 12. Electrical and electronic technicians 13. Occupational therapists 14. Surveyor helpers 15. Credit clerks, banking and insurance 16. Physical therapists 17. Employment interviewers 18. Mechanical engineers 19. Mechanical engineering technicians 20. Compression and injection mold machine operators, plastics Average Female: Male Percentage Percentage Weekly Earnings Ratiob Growtha Femaleb Earningsb (x 100) 96.8 94.3 85.3 76.9 75.8 71.7 67.8 65.3 63.9 63.5 62.2 60.7 59.8 58.6 54.1 53.6 52.5 52.1 51.6 50.3 NA NA 25.1 28.4 63.2 5.6 82.7 3.5 17.8 63.2 NA 9.7 NA NA NA 67.3 48.7 2.5 17.8 NA NA NA $519 422 NA NA 76.9 73.6 260 67.8 327 209 549 348 75.3 260 67.8 NA NA 387 NA NA NA 305 402 540 348 NA NA NA 67.3 64.3 75.3 NA NA Total 25.0 39.5 $289 64.7 NOTES: NA = not available. = not shown because base is less than 50,000. a Includes only detailed occupations with 1982 employment of 25,000 or more. b Approximate due to use of different occupational classifications in sources. Data are based on 1981 annual averages; earnings referto weekly earnings of full-time wage and salary workers in occupa- tions of 50,000 or more. SOURCES: Rytina(1982:Table DandSilvestriet al. (1984:Table 1). are expected to add 1.6 million jobs to the economy by 1995; the jobs listed in Table 11 will add 9.2 million. WOMEN'S STATUS IN THE LABOR MARKET Trends in Job Segregation As is apparent from the foregoing discussion, particular occupations are closely identified with and often performed almost exclusively by workers of one sex or the other. The distribution of women across major occupational
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY 153 TABLE 1 1 The 20 Occupations With Largest Projected Job Growth Between 1982 and 1995 Occupation 1. Building custodians 2. Cashiers 3. Secretaries 4. General clerks, office 5. Sales clerks 6. Nurses, registered 7. Waiters and waitresses 8. Teachers, kindergarten and elemental 9. T'uck drivers 10. Nursing aides and orderlies 11. Sales representatives, technical 12. Accountants and auditors 13. Automotive mechanics 14. Supervisors of blue-collar workers 15. Kitchen helped 16. Guards and doorkeepers 17. Food preparation and service workers, fast food restaurants 18. Managers, store 19. Carpenters 20. Electrical and electronic technicians Total Average Female: Male Percentage Percentage Weekly Earnings Ratioh Growths Females Earningsh (x 100) 27.5 47.4 29.5 29.6 23.5 48.9 33.8 37.4 26.5 34.8 29.3 40.2 38.3 26.6 35.9 47.3 36.7 30.1 28.6 60.7 25.0 14.6 85.1 99.3 83.5 60.3 95.8 85.2 82.2 2.1 84.3 NA 39.7 .7 10.5 30.5 12.8 85.0 38.2 1.4 9.7 39.5 219 168 230 192 178 332 150 322 314 172 NA 379 285 394 135 232 141 300 325 387 $289 83.6 92.0 60.3 72.0 82.2 82.2 NA 71.2 64.2 90.7 57.0 NOTES: NA = not avai table. = not shown because base is less than 50 000. a Includes only detailed occupations with 1982 employment of 25,000 or more. b Approximate due to use of different occupational classifications in sources. Data are based on 1981 annual averages; earnings refer to weekly earnings of full-time wage and salary workers in occupa- tions of 50,000 or more. SOURCES: Rytina (1982:Table 1) and Silvestr~ et al. (1984:Table 1). categories is shown in Table 12 for 1970 and 1980. Although some changes have occurred (notably among executives and managers), certain categories remain highly sex-typed, e.g., clerical, other services, private household, craft workers, and laborers. An overall index of the sex segregation of occupations has been developed that measures the minimum proportion or percentage of persons of either sex who would have to change to an occupation in which their sex is underrepre- sented in order for the occupational distributions of the sexes to be identical (Reskin and Hartmann, 19851. Segregation indices calculated since 1972
154 CAIN TABLE 12 Percentage Distribution of Women in Major Occupational Groups, Change, and Rate of Growth in Female Representation, 1970-1980 Change (1970- Rateof Major Occupational Group 1970 1980 1980) Grown Executive, administrative, managerial 18.5 30.5 + 12 .65 Professional specialty 44.3 49.1 +4.8 .11 Technicians 34.4 43.8 + 1.8 .04 Sales 41.3 48.7 +7.4 .18 Administrative support, including clerical 73.2 77.1 +3.9 .05 Private household 96.3 95.3 - 1.0 .01 Protective service 6.6 11.8 +5.2 .79 Other service 61.2 57.2 -4.0 .07 Farming, forestry, fishing 9.1 14.9 +5.8 .64 Precision production, including craft 7.3 7.8 +O.S .07 Machine operators 39.7 40.7 + 1.0 .03 Transportation 4.2 7.8 +3.6 .86 Handlers, laborers 17.5 19.8 +2.3 .13 NOTE: Both 1970 and 1980 data are coded to the 1980 census occupational classification. SOURCE: Rytina and Bianchi (1984:Table 2). across detailed occupations show a decline from 68.3 to 61.7 (Reskin and Hartmann, 19851. Belier and Han (1984) developed several sets of segrega- tion projections, all of which extrapolate 1970 trends to the 1980s. By most of their estimates, the segregation index will continue to drop, to about 60.0 by 1990. Their most optimistic forecast, however, shows a drop to 41.3 . The magnitude of these declines can be better understood by comparing them with the decline of 25 percentage points that Blau and Hendricks (1979) estimated would occur if past hiring remained in place but new hiring was fully integrated or blind with regard to sex. The value of the segregation index is affected by two factors: the number of people employed in different sex-typed occupations (the mix effect) and the sex composition of workers within an occupation (the composition effect). Blau and Hendricks (1979) found an increase in segregation during the 1950s, which was primarily due to the considerable growth of female- dominated occupations, notably clerical workers and professions such as teaching and social work. During the 1960s the segregation index declined as the growth of female-dominated jobs slowed and the share of workers (predominantly men) entering jobs held largely by the opposite sex increased. Using Current Population Survey data, Belier (1982) found that these trends continued during the 1970s and accounted for much of the decline in segregation in this period. A recent study by Rytina and Bianchi (1984) examined changes in the sex composition of occupations using
PROSPECTS FOR PAY EQU17~Y IN A CHANGING ECONOMY 155 results of the 1970 and 1980 censuses. Their findings corroborate and expand those of Belter and Han ( 1984) . They too found a decline in segrega- tion during the 1970s, which resulted from three factors: (1) a substantial drop in male-intensive occupations those in which 20 percent or less of workers were women; (2) a modest rise in sex-neutral occupations those employing 21 to 59 percent women; and (3) no increase in the number of female-dom~nated occupationsthose with 60 percent or more female incumbents. Thus, in terms of employment, both men and women were more likely to be in neutral or integrated occupations by 1980. Neither men nor women, however, were significantly more likely to be in occupations dominated by the opposite sex, so that the occupations that were most closely identified with one sex or the other remained that way over the decade. Thus, there was little or no change in the sex composition of jobs in major female-intensive categories such as clerical and private household workers, nor in male-intensive categories such as transportation, laborer, or craft worker. The single greatest change occurred among managers, a category in which women increased from 18 to 31 percept oftotal employment. For both men and women, movement into sex-neutral occupations coincided with declines in employment in traditionally sex-dominated jobs. Several features of this apparent decrease in job segregation are worth noting. First, change appears to be greatest in occupations et the upper end of the socioeconomic spectrum among managers, professionals, and techni- cians occupations that employ a relatively small proportion of the total labor force, either male or female. Second, as Rytina and Bianchi (1984) point out, within sex-neutral or integrated occupations, segregation is still possible within subspecialties or in jobs within firms. For example, in a study of several hundred establishments, Bielby and Baron ( 1984) found that job classifications were almost completely segregated. They also found that the level of segregation remained constant over the 20-year period from 1959 to 1979, during which equal employment opportunity laws were passed and visibly enforced. In summary, although there have been changes in Me sex composition of individual occupations, recent trends and projec- tions for the future reveal only modest change in and persistent high levels of job segregation. On a more optimistic note, there are a number of considerations that would lead one to predict a continued decrease in job segregation. It appears that younger women, who are new or recent entrants to the labor force, are taking the lead in moving into nontraditional occupations (Belier, 1982; Reskin and Hartmann, 19851. As may already be occurring, the relative shortfall of young male workers, coupled with the rising work aspirations of young women, could enhance employers' receptivity to hiring women for nontra-
156 CAIN ditional jobs. Employers may also be less likely to rely on outmoded stereo- types to screen and place women in segregated jobs. The higher pay, on average, of nontraditional or male-dominated jobs may also make them increasingly attractive to women who maintain families or who contribute a large share of household income in dual-worker couples. Women's acquisi- tion of more work experience, education, and training should result as well in greaterjob market savvy, enabling them to better identify and take advan- tage of favorable job opportunities. The highly visible progress of women in professional occupations also enhances the position of women by providing role models and a sense of empowerment, thereby initiating an effect that would ultimately benefit women at all levels of the occupational structure. The projected slowdown in male-dominated manufacturing jobs and in female-dominated clerical jobs could also be conducive to greater integra- tion if future trends follow the patterns of the 1970s. During that decade, decline in traditional, segregated occupations coincided with movement into integrated or sex-neutral ones rather than leading to the creation of new male- or female-dominated bastions. Projected changes in the age and race or ethnic composition of the future labor force also have potential implications for job segregation. Although there is considerable controversy over the effectiveness of existing aff'rma- tive action programs, which set goals and timetables for the recruitment and hiring of protected classes of workers, the existence of these programs serves, at a minimum, to legitimate the right of women to enter nontradi- tional occupations. These programs typically target workers for ent~y-level jobs and accordingly have been aimed at younger workers. The declining numbers of younger workers might change their focus in favor of older, returning workers, among whom women would figure prominently. But the changing demography of the labor force could also lead to stagnation or erosion of women's position in the labor market. As the number of younger, minority workers increases through immigration and as a result of race differentials in fertility, women's concerns (which are neither mutually exclusive nor identical with those of minorities) may lose visibility. A simi- lar competition may also arise between the interests of women and minority workers and those of older white male workers. Age discrimination legisla- tion passed in 1967 sets the stage for an increasing number of lawsuits pertaining to layoffs and early retirements. If these cases are successful, worker turnover will be slowed and white men stand to benefit at the expense of wider opportunities for younger or less experienced workers. The projected high rates of growth of many female-dominated occupa- tions might also tend to slow the decline in job segregation, since women could be expected to seek out or be recruited for occupations in which they
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY 157 had already established entry. This tendency might be especially pro- nounced among mothers of young children and reentering women (who will account for most of the upcoming increase in female labor force participa- tion), because they are less knowledgeable about the job market and poten- tially more constrained in their choices by virtue of family responsibilities. This possible trend might be offset, however, by a movement of female heads of household into higher-paying, male-dominated jobs. Another factor that would tend to maintain segregation at current levels is the increasingly smaller size of cohorts entering the labor force. Because of their smaller numbers, younger cohorts would have to enter nontraditional occupations at extremely high rates in order to prompt significant changes in the sex composition of given occupations. Younger women are pacesetters; however, the majority of them still aspire to and enter female-dominated jobs (Marini and Brinton, 1984~. Contributing to this more pessimistic prognosis is the fact that much of the decline in job segregation to date has reflected the activities of an extremely large cohort in the early stages of their work lives. This is a period of especially great experimentation and job mobility. Since job and occupation switching decline with age (Rytina, 1983), the rate of movement into and out of occupations could slow as the baby boom cohort settles down to particular jobs or careers, thus slowing changes in the sex composition of jobs. In view ofthese tendencies pointing in opposite directions, it is difficult to predict future changes in the degree of job segregation. On balance, demo- graphic factors suggest a slowing in the rate of decrease in segregation that was observed in the 1970s. Yet a strong women's movement or government enforcement effort coupled with economic growth could certainly contrib- ute to further substantial decline. The Wage Gap The persistence of a considerable disparity in the pay of full-time, year- round male and female workers has been labeled the "wage gap." Typically the wage gap is measured as the ratio of female-to-male median annual or hourly earnings, or in terms of the actual dollar difference between them. Since 1955, women's earnings relative to those of men have fluctuated at around 60 percent (U.S. Department of Labor, 1982:Table 12~. In 1982 the ratio had moved to 62.0 percent, which represented a difference in annual earnings of $7,976 (Norwood, 1984~. In constant dollars, from 1955 to l9X1, the earnings gap increased by 60 percent, from $1,911 in 1955 to $3,032 in 1981 (U.S. Department of Labor, 1982~. As Norwood (1984) notes, the size of the wage gap shrinks as additional considerations such as occupation, education, work experience, and age are taken into account.
158 CAIN Thus, for example, looking at 1981 salary levels for a group of narrowly defined white-colIarjobs, Sieling (1984) found that the earnings gap was in the O to 16 percent range, smallest within the same establishments, and not always in men's favor. O'Neill (1983) argues that trends in the wage gap can be explained in part by the changing composition of the female labor force and concomitant changes in women's standing relative to men on productivity-related charac- teristics. Thus, with theinfluxofwomenintothelaborforcein the 1960s and 1970s, the level of education of employed women as well as their work experience declined and the wage gap widened. The narrowing of these differentials did not appear to influence the wage gap until about 1979; since that time there has been a slight improvement in women's favor. O'Neill notes that the decline in the gap was greatest for workers ages 25 to 34. She concludes that wage differentials are likely to narrow in the next decade as young women work more continuously, increase their education, and raise their work expectations. Another study, which projected men's and women's pay to the year 2000, concurred with this assessment (Smith and Ward, 1984~. By estimating the earnings of future female labor force participants and averaging them with those of women already in the labor force, they arrived at a projected average wage for all women. Their findings indicate that if current labor force trends were to continue, women would earn 74 percent of what men do by the end of the century narrowing but not closing the wage gap from approximately 40 to 25 percent. In contrast to these optimistic assessments, a recent study by Green (1983) found that the wage gap widened from 1970 to 1980 for recent labor force entrants, despite the fact that the productivity-related characteristics of males and females became more similar over this period. Green also fol- lowed the progress of the cohort of workers who entered the labor force in 1970 by examining outcomes for a panel of experienced workers in 1980. Among whites he found that the male-female wage gap widened from 14 to 32 percent. Thus, after 10 years of potential work experience, the wage gap of the experienced 1970 cohort approximated the overall gap of about 38 percent for the entire labor force. Because there was no direct measure of work experience in Green's data, some amount of the observed gap could be due to unmeasured breaks in women's actual work experience, particularly that of married women. Yet the trends in women's labor force participation reviewed earlier suggest that this cohort had relatively high levels of activity. Green's results suggest instead that there had been a substantial increase in sex discrimination, especially against white women, over the last decade. By his estimates, discrimination (including discrimination in wages and - ~~
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY 159 occupational placement) could have accounted for as much as 79 percent of the gap in 1980, compared with 43 percent in 1970. Green's results also indicate that occupational segregation was more important in explaining wage differences for ent~y-level workers, while wage discrimination was more important among experienced workers. He speculated that the lack of progress by white women (compared with that of black men, who did make wage gains relative to white men) may be due to heightened job competition among women and minority men and to the role of equal employment and affirmative action programs in awarding jobs. Studies of male-female earnings differences that have been conducted over the last 15 years also conclude that some portion (as much as 30 to 40 percent) of the wage gap is due to discrimination (see Treiman and Hart- mann, 1981, for a review). Although there is considerable controversy surrounding the use of residual or unexplained variation to estimate its impact~see, for example, Cole 1979), to the extent that discrimination exists (and may be increasing), it augurs badly for women's achievement of equal pay. Because the more women there are in an occupation the less it pays (Treiman and Hartmann, 1981), job segregation is a significant factor in maintaining wage differentials. Thus, the mixed prospects for a decline in segregation do not bode well for women's progress in closing the wage gap. Moreover, occupational projections indicate that a large number of new jobs will be in what have been historically low-paying sectors, e.g., retail sales, maintenance work, and lower-level health care jobs. Thus, the weekly earnings of about half the large-growth jobs in Table 11 fall below the labor- force-wide average. A recent study (Bluestone et al., 1984) concludes that by the year 2000 most jobs will be in sectors of the economy that currently have average annual earnings of less than $12,500 and there will be a substantial decline in jobs in industries paying more than $22,000 per year. Unless women are able to move in significant numbers out of the jobs in which they are now concentrated, they are likely to be found disproportion- ately in these low-paying sectors. At the same time, women's inroads in professional occupations do signal substantially improved earnings pros- pects for a relatively small, elite group. In addition, for women workers in other occupational groups, increased unionization may improve earnings (Freeman and Leonard, 19841. A large portion of the pay gap (another 30 to 40 percent) is due to differ- ences in the productivity-related characteristics of men and women in the same occupations. The prospects for nan owing these male-female differen- tials appear better, as the studies cited earlier by O'Neild (1983) and Smith and Ward (1984) indicate. Women's accrual of more on-thejob training, tenure within a firm, and seniority should result in higher eanungs, assum-
160 CAIN ing that men and women receive similar rates of return on these factors. To the extent that women's rates of return are lower than those of men (Roos, 1981), however, the wage gap will not close, at least not as much as pre- dicted. Even under the most optimistic assumptions, it appears that a sizable wage gap will persist through the end of the century. Wage gap comparisons, however, typically take into account only full-time, year-round workers. Because part-time work is expected to increase throughout the economy and because women's rate of part-time work is likely to remain much higher than that of men, the real wage gap between men and women is indeed much larger than the foregoing estimates suggest. Although part-time work appears to be attractive to women who are juggling family responsibilities, it appears to offer little or no wage growth (Corcoran et al., 19841. More insidiously, women's higher rates of involuntary part-time work and their increasing propensity to hold multiple jobs suggest that women are being channeled into, rasher then choosing, this option. The increased presence of women in the labor force has led to several developments that, if successful, could hasten a closing of the wage gap. Foremost among these is the push for comparable worth or pay equity. Comparable worth strategies seek to raise the wages of traditional female- dominated jobs. Comparable worth advocates argue that pay should reflect the skill, effort, and responsibility entailed in jobs and that jobs with similar features should be paid the same whether they are performed by women or men. Comparable worth has made considerable, rapid progress, especially in the public sector. It has been implemented by several state.and local governments and is increasingly the basis of collective bargaining and orga- nizing efforts. Numerous studies, preliminary to establishment of a pay equity policy, are also under way (Dean et al., 19841. Moreover, several recent court decisions have encouraged comparable worth claims. The Supreme Court's 1981 ruling in Gunther v. County of Washington, Oregon (101 S. Ct. 2242) opened the door to such claims, because it found that claims of wage discrimination in different jobs could be heard under Title VII; the Court specifically denied, however, that it was endorsing compara- ble worth. A 1983 U.S. district court ruling against Washington State (which is still under appeal) upheld the validity of a comparable worth strategy and particularly the use of a job evaluation plan by the state to determine compa- rablejobs. Thus far, successful comparable worth strategies have succeeded in raising wages among female-dominated jobs in the clerical, health, social work, and nursing fields. When earnings growth for men is examined by age or over the life cycle, it is found to be greatest during the late 20s and 30s, a period that coincides with the years of women's peak childbearing and rearing. To the extent that
PROSPECTS FOR PAY EQUITY lN A CHANGING ECONOMY ~ ~ ~ ~ . . ~ . . 161 women-e ~anorrorce attachment during this time can be enhanced, the payoff in increased earnings should be especially pronounced. Thus, other devel- opments, such as the increased interest in child care and alternative work arrangements, also have potentially favorable implications for women's pay. Access to more and better child care and the availability of schedules that are conducive to meeting family demands should increase women's ability to go to work and stay on the job. The strength of these developments in the face of a political environment in which equal employment goals have been neglected attests to the urgency of the needs of women workers. Their own activism and organizing efforts on these fronts are perhaps the greatest basis for optimism that the wage gap will continue to narrow. IMPLICATIONS FOR FUTURE RESEARCH In light of the foregoing discussion, a research agenda to better illuminate these issues might include the following topics: 1. Sophisticated modeling and simulation exercises to better assess future directions in sex segregation and pay equity in light of demographic and economic changes. Such efforts need to explicitly incorporate (a) the age structure both ofthe labor force and of occupations; (b) age differences injob and occupational mobility; (c) cohort size and race and ethnic composition; (d) effects of changes in sex composition on occupational wages; and (e) training and skill needs (the last two factors were not dealt with in this paper but need to be taken into account in any large-scale modeling effort). 2. Inquiry into the overall and possibly disparate impact of technological change on men and women's occupational distribution and pay. 3. Better understanding of the nature of work in newly emerging indus- tries, with a special focus on unionizing efforts and on those seeking to introduce innovative forms of workplace organization. 4. Examination of the economic interests of and prospects for coalition or competition between women and minorities (male and female, foreign- and native-born). 5. An inquiry into the impact of child care and alternative work arrange- ments on labor force attachment and job choice, because most of the pro- jected increase in women's labor force participation will come from the mothers of young children. 6. Case studies to investigate the so-called sex-neutral occupations (which account for much of the recent decline in sex segregation) to better understand some of the factors that appear to promote the entry of bow men and women.
162 CAIN 7. Now that comparable worm is beginning to be implemented, albeit still on a limited basis, evaluation of its scope and impact in increasing Me wages of particular "women's" jobs and of its implications for levels of employment in affected jobs and forge sex composition of workers in ~em. REFERENCES Andreassen, A.J., N.C. Saunders, and B.W. Wu 1984 Economic outlook for the 1990's, three scenarios for economic growth. Pp. 9-21 in Bureau of Labor Statistics, Employment Projectionsfor 1995. Washington, D.C.: U.S. Department of Labor. Belter, A. 1982 Occupational segregation by sex: Determinants and changes. Journal of Human Resources 17(3):371-372. Belier, A., and K. Han 1984 Occupational sex segregation: Prospects for the 1980s. Pp. 91-114 in B. Reskin, ea., Sex Segregation in the Workplace: Trends, Explanations, Remedies. Committee on Women's Employment and Related Social Issues. Washington, D.C.: National Acad- emy Press. Bielby, W., and J. Baron 1984 A women's place is with other women: Sex segregation within organizations. Pp. 27-55 in B. Reskin, ea., Sex Segregation in the Workplace: Trends, Explanations, Remedies. Committee on Women's Employment and Related Social Issues. Washington, D.C.: National Academy Press. Blau, F.D., and W.E. Hendricks 1979 Occupational segregation by sex: Trends and prospects. Journal of Human Resources 14(2): 197-210. Blau, F.D., and L.M. Kahn 1981 Race and sex differences in quits by young workers. Industrial and Labor Relations Review 34(4):563-577. Bluestone, B., B. Harrison, and L. Gorham 1984 Storm Clouds on the Horizon: Labor Market Crisis and Industrial Policy. Economic Education Project, Brookline, Mass. Bureau of Labor Statistics 1982a Labor Force Statistics Derived From the Current Population Survey: Databook. Vol. . 1. Washington, D.C.: U.S. Department of Labor. 1982b BLS Handbook ofMethods. Washington, D.C.: U.S. Department of Labor. 1983a Women at Work: A Chartbook. Washington, D.C.: U.S. Department of Labor. 1983b Marital and Family Patterns of Workers: An Update. Washington, D.C.: U.S. Depart- ment of Labor. 1984a Employment Projectionsfor1995. Washington, D.C.: U.S. Department of Labor. 1984b Families at Work: The Jobs and the Pay. Washington, D.C.: U.S. Department of Labor. Chiswick, B.R. 1982 Immigrants in the U.S. labor market. Annals of the American Academy of Political and Social Science 460(1):64-72. Cole, J. 1979 Fair Science: Women in the Scientific Community. New York: Columbia University Press.
PROSPECTS FOR PAY EQUITY IN A CHANGING ECONOMY 163 Corcoran, M., G. Duncan, and M. Ponza 1984 Work experience, job segregation, and wages. Pp. 171-191 in B. Reskin, ea., Sex Segregation in the Workplace: Trends, Explanations, Remedies. Committee on Wom- en's Employment and Related Social Issues. Washington, D.C.: National Academy Press. Dean, V., J.A. Gone, M. Klaw, and D. Mitchell 1984 Who's Working for working Women? National Committee on Pay Equity, Washington, D.C.. Finch, J.L. 1983 Worklife estimates should be consistent with labor force rates. Monthly Labor Review 106(6):34-36. Freeman, R.B., and J.S. Leonard 1984 Union Maids: Unions and the Female Workforce. Unpublished paper, Conference on Genderin the Workplace, Brookings Institution, Washington, D.C. Fullerton, H.N., and J. Tschetter 1984 The 19951aborforce: A second look. Pp. 1-8 in Bureau of Labor Statistics, Employment Projectionsfor 1995. Washington, D.C.: U.S. Department of Labor. Green, G.W. 1983 Wage Differentials for Job Entrants, by Race and Sex. Unpublished Ph.D. dissertation, Department of Economics, George Washington University. Haber, S.E., E.J. Lamas, and G. Green 1983 A new method for estimating job separation rates by sex and age. Monthly Labor Review 106(6):20-27. Johnson, B.L., and E. Waldman 1984 Most women who maintain families receive poor labor market returns. Pp. 15-19 in Bureau of Labor Statistics, Families at Work: The Jobs and the Pay. Washington, D.C.: U.S. Department of Labor. Keely, C.B. 1979 U.S. Immigration: A Policy Analysis. New York: Population Council. Leon, C.B. 1982 Occupation winners and losers: Who they were during 1972-80. Monthly Labor Review 105(6): 18-28. Levitan, S.A., and C.M. Johnson 1982 The future of work: Does it belong to us or the robots? MonthlyLaborReview 105(9): 10- 14. Lloyd, C.B., and B.T. Niemi 1979 The Economics of Sex Differentials. New York: Columbia University Press. Marini, M.M., and M.C. Brinton 1984 Sex typing in occupational socialization. Pp. 192-232 in B. Reskin, ea., Sex Segrega- tion in the Workplace: Trends, Explanations, Remedies. Committee on Women's Employment and Related Social Issues. Washington, D.C.: National Academy Press. Mellor, E.F. 1984 Investigating differences in weekly earnings of women and men. Monthly Labor Review 107(6): 17-28. No~wood, J. 1984 Working Women and Public Policy. Address presented at the National Conference on Women, the Economy and Public Policy, Washington, D.C. O'Neill, J.A. 1983 The Trend in the Sex Differential in Wages. Working paper, Urban Institute, Washing- ton,D.C.
164 CAIN Personick, V.A. 1984 The job outlook through 1995: Industry output and employment projections. Pp. 22-34 in Bureau of Labor Statistics, Employment Projections for 1995. Washington, D.C.: U.S. Department of Labor. Reskin, B.F., and H.I. Hartmann, eds. 1985 Women's Work, Men's Work: Sex Segregation on the Job. Committee on Women's Employment and Related Social Issues. Washington, D.C.: National Academy Press. Riche, R.W., D.E. Hecker, andJ.O. Burgan 1983 High technology today and tomorrow: A small slice of the employment pie. Monthly Labor Review 106(11):50-58. Roos, PA. 1981 Sex stratification in the workplace: Male-female differences in returns to occupation. Social Science Research 10(3): 195-224. Rytina, N.F. 1982 Earnings of men and women: A look at specific occupations. Monthly Labor Review 105(4):25-31. 1983 Occupational changes and tenure, 1981. Pp. 4-34 in Bureau of Labor Statistics, Job Tenure and Occupational Change, 1981. Washington, D.C.: U.S. Department of Labor. Rytina, N.F., and S.M. Bianchi 1984 Occupational reclassification and changes in distribution by gender. Monthly Labor Review 107(3): 11-17. Sekscenski, E.S. 1981 Women's share of moonlighting nearly doubles during 1969-79. Pp.36-39 in Bureau of Labor Statistics, Multiple Jobholders in May 1979. Washington, D.C.: U.S. Depart- ment of Labor. Serrin, W. 1983 "High tech" is no job panacea, experts say. New York Times (Nov. 18). 1984 Shifts in world put white men in the minority. New York Times (July 31). Sieling, M.S. 1984 Staffing patterns prominent in female-male earnings gap. MonthlyLaborReview (6):29- 33. Silvestri, G.T., J.M. Lukasiewicz, and M.E. Einstein 1984 Occupational employment projections through 1995. Pp. 35-47 in Bureau of Labor Statistics, Employment Projections for 1995. Washington, D. C.: U. S. Department of Labor. Smith, J.P., and M. Ward 1984 Women 's Wages and Work in the Twentieth Century. Santa Monica, Calif.: Rand Corpo- ration. Smith, R.E., ed. 1979 The Subtle Revolution. Washington, D.C.: Urban Institute. Smith, S.J. 1982 New worklife estimates reflect changing profile of labor force. Monthly Labor Review 105(3): 15-20. Labor force participation rates are not the relevant factor. Monthly Labor Review 106(6):36-38. Treiman, D.J., and H.I. Hartmann, eds. 1981 Women, Work, and Wages: Equal Payfor Jobs of Equal Value. Committee on Occupa- tional Classification and Analysis. Washington, D.C.: National Academy Press. 1983
PROSPECTS FOR PAY EQUITY IN A CHANGl~G ECONOMY U.S. Department of Labor 1982 Equal Employment Opportunitiesfor Women: U.S. Policies. Washington, D.C. Department of Labor. 165 : U.S. Viscusi, W.K. 1980 Sex differences in worker quitting. Review of Economics and Statistics 62(3):388-398. Waite, L.J. 1978 Projecting female labor force participation from sex-role attitudes. Social Science Research 7(4):299-318. Waldman, E. 1984 Labor force statistics from a family perspective. Pp. 1-5 in Bureau of Labor Statistics, Families at Work: The Jobs and the Pay. Washington, D.C.: U.S. Department of Labor.