This file was prepared for electronic distribution by the inforM staff. Questions or comments should be directed to inform-editor@umail.umd.edu. The Economic Status of Black Women: An Exploratory Investigation Staff Report United States Commission on Civil Rights October 1990 Part IV Conclusion Chapter 10 Conclusion This report contains both heartening and disheartening findings with respect to black women's economic status. On the one hand, the pay of black women has increased substantially, both relatively and absolutely, over the past half-century. A primary cause of black women's increased relative pay appears to have been a substantial decline in the effects of racial discrimination in the labor market. In 1940 black women's hourly wages were barely one-half those earned by comparable white women. Today, black women earn roughly 90 percent as much as comparable white women. The occupational distributions of black and white women with similar characteristics have undergone an equally large convergence between 1940 and the present. On the other hand, despite these increases in relative pay and occupational status, black women still earn less than white women, and black women's economic status continues to be far below white women's. Black women's average family income is less than two-thirds that of white women. Black women are three times more likely to have family incomes of less than $10,000, and seven times less likely to have family incomes of more than $60,000. Black women's median family net worth is $8,335, less than one-fifth as high as white women's, which is $45,659. Black women are five times more likely to be in poverty, five times more likely to be on welfare, and three times more likely to be unemployed than white women. Many factors, some related to current labor market discrimination and others not, combine to lower black women's economic status. Differences in the family structures of black and white women are a major reason why black women's economic circumstances are worse than white women's. Whereas roughly two-thirds of white women are married, roughly the same proportion of black women are not married. Unmarried black women are considerably more likely to have children than their white counterparts. For instance, 44 percent of black women who have never been married have at least one child under 18, compared with 6 percent of white women. These differences in family structure mean that black women are more likely than white women to be the only earners in their families, and if they are, to be responsible for children. A second important factor lowering black women's economic status is the lower incomes of other family members, especially the lower earnings of their husbands. Black women's husbands earn an average of $18,013 per year, much less than white women's husbands, who earn $26,417 on average. Forty-five percent of black wives have husbands earning less than $15,000 per year, compared with 29 percent of white wives, and fewer than 8 percent of black wives have husbands earning over $45,000 per year, compared with over 20 percent of white wives.(1) It should be remembered in this regard that current racial discrimination in the labor market may be partially responsible for black men's lower earnings.(2) Because other family members contribute relatively less to family income, black women's earnings constitute a higher proportion of their family income than white women's. Overall, black women contribute one-third of their family income, whereas white women contribute one-fourth.(3) Consequently, factors that lower black women's earnings can have substantial adverse effects on their economic status and on that of their families. Factors that lower black women's earnings can be divided into those that affect their hourly wages and those that affect the number of hours they work. Part II of this report concentrates on analyzing the former, and part III focuses on the latter. This chapter summarizes the major results of parts II and III and evaluates the effects of labor market discrimination on black women's economic status. It also develops an agenda for future research to improve understanding of the nature and effects of labor market discrimination. Black Women's Economic Status: A Summary of Results This report has traced the gap in wages earned and occupations held by black and white women from 1940 to the present and assembled statistical evidence pertaining to the sources of the gap. The report has also considered other factors affecting black women's relative economic status in the 1980s. Here, the evidence presented in earlier chapters is summarized and assessed for its implications concerning the nature and degree of race-based labor market discrimination against black women. Trends in Wages and Occupations Evidence on the trends in wages and occupations from the U.S. Censuses of Population and the Current Population Surveys leaves no doubt that black women's relative wages increased substantially and that black-white occupational differences narrowed considerably over the period from 1940 to the present. In 1940 black women earned only 40 percent as much per hour as white women. In the 1980s, black women have achieved something close to wage parity with white women: depending on the data source used, black women's relative wage has risen to between 89 and 100 percent.(4) Paralleling the increase in black women's relative wage has been an equally large improvement in the relative occupational status of black women. In 1940 close to 60 percent of black women were domestic servants, and another 11 percent were farm laborers. Only 9 percent of white women worked in these two occupational categories combined. Black women were barred from many blue-collar and service sector jobs (other than in private households) and from virtually all clerical, sales, and professional jobs (except for teaching). By 1980 black women had left their jobs as farm laborers and domestic servants and made substantial inroads into all of these occupations. Whereas the occupational distributions of black and white women in 1940 were almost completely distinct, they were very similar in 1980. Because black women in 1940 were heavily represented among farm laborers and domestic servants, both jobs in which some compensation may be received ln the form of in-kind payments, black women's 1940 relative wage may have been somewhat larger in real terms than the 40 percent figure cited above, which was calculated on the basis of monetary wage payments only. Since black women had largely left these jobs by 1980, the 1940-80 growth in black women's relative wages may have been somewhat less than appears when only monetary wages are taken into consideration. Nevertheless, black women's wages have undoubtedly increased substantially relative to white women's since 1940. Younger black women experienced stronger relative improvements in wages and occupations than older black women, and the pace of the improvement also varied by region and by level of education. The South lagged behind the rest of the country. Whereas black women's relative wage and occupational status began to rise in the rest of the country before 1960, it was after 1960 before any noticeable improvement occurred in the South. Similarly, black women with low levels of education saw their relative wages and occupational attainment improve only after more educated black women did. This occurred partly because less educated black women tended to be older than their more educated counterparts and because they were more likely to live in the South. Despite the major improvements in black women's wages and occupations since 1940, there is strong evidence suggesting that the wages and occupational status of black women continue to lag behind those of white women. Wage data for the 1980s taken from the Current Population Surveys suggest that the relative hourly wage of black women reached 89 percent in 1980 and did not continue to increase during the 1980s. When only full-time year-round workers are considered, the relative wage of black women appears even to have fallen slightly during the 1980s. In addition to having lower wages than white women, black women are still less likely than white women to be in professional, managerial, clerical, and sales occupations, and more likely to work as operatives, service workers, and domestic servants. Black-white wage and occupational differentials continue to be wider in the South than in the rest of the country, and wider for older and less educated black women. Accounting for Trends in Wages and Occupations To investigate the reasons for trends in black women's relative wages and occupational status, this report took into account black- white differences in factors known to affect women's wages and occupations, such as education. work experience, marital status, and the presence of children.(5) Census data for the years 1940 to 1980 and data from the Current Population Surveys and the Survey of Income and Program Participation for the 1980s show that even after racial differences ln characteristics were taken into account, black women on the whole earned less than comparable white women(6) during the entire period under consideration. In 1940 black women earned barely 50 percent as much as comparable white women. This ratio increased to 65 percent in 1960 and to somewhere between 90 and 100 percent in the 1980s. This report also evaluated the separate effects of the following individual factors on black women's relative wages: education, regional distribution, urban or rural residence, part time or full time work status, children, marital status, and age. Although black and white women differed substantially in all of these characteristics, only racial differences in education were found to lower noticeably black women's relative wages. Black women's lower educational levels did account for a small portion of the black-white wage gap throughout the period.(7) The report has found that racial differences in occupations played a central role in limiting black women's relative wages over the entire period. Racial differences in occupational distributions alone accounted for most of the large black-white wage gaps in 1940 and 1960. Both the wage gap and occupational differences have diminished considerably since 1960, but racial differences in occupations continue to account for a large share of the remaining wage gap. (9) Diminishing occupational differences between black and white women also appear to have been a central factor in the improvement of black women's relative pay since 1940. The black-white wage gap declined in tandem with racial differences in occupations. Occupational differences diminished slightly between 1940 and 1960, and so did the black-white wage gap. Between 1960 and 1980, occupational differences diminished substantially, and the black-white wage gap did the same. Throughout the period. occupational differences diminished first among educated women and outside of the South-- and so did the black-white wage gap. Chapter 6 finds that almost one-half of the increase in black women's relative wage between 1940 and 1980 was due to their improved relative occupational status.(10) The increasing similarity of black and white women's educational levels, regional distributions, and proportions living in urban areas also each accounted for a small part of the increase.(11) Given the central importance of occupations in determining black women's relative wages, it is essential to understand the reasons for black women's historically low occupational status and for the improvements that came about, before 1960 for some black women but for the most part, after 1960. The analysis reported in chapter 5 finds that racial differences in education, age, region, and urban-rural distribution can account for only small portions of the racial differences in occupations in the years 1940, 1960, and 1980. (12) One factor that may have limited black women's occupational opportunities (and also their wages) throughout the period is the lower quality of the education received by black women. As noted in chapter 4, black women who were educated in the first half of this century were likely to have received considerably fewer days of schooling than white women who reported the same grade level. Up until the 1960s, black women were educated, for the most part, in poorly funded, racially segregated schools. There is substantial evidence that even in the 1980s black women have lower educational achievement (as measured by test scores on various achievement tests) than white women who have completed the same number of years of school.(13) Occupational differences between black and white women had narrowed considerably by 1980. Nevertheless, in 1980 black women continued to be underrepresented in certain occupations, most notably clerical occupations.(14) Since clerical work does require a fair amount of schooling, it is possible that lower quality education impedes black women's entry into clerical jobs. Differences in the quality of education are not a full explanation for the historical differences between the occupations of black women and those of comparable white women. For instance, differences in the quality of education are unlikely to explain southern black women's almost total exclusion from jobs as operatives in the textile industry, since these jobs do not require much education. Differences Across Age, Region, and Education Levels As indicated above, chapters 6 and 7 found that younger black women generally have fared better relative to similarly qualified white women than have older black women. In the 1980s black women over 40 earn only 88 percent as much as comparable white women, whereas black women under 40 earn 94 percent as much as comparable white women. Older black women's lower relative earnings appear to be because they are in lower status occupations relative to their white counterparts. This result suggests that in addition to the effects of current discrimination, older black women have not overcome the effects of past labor market discrimination. They grew up in a time when educational and occupational opportunities for black women were severely restricted. Thus, past discrimination reduces older black women's economic status today, because it has a lasting impact on their educational attainment and occupational status. Similarly, southern black women have fared worse relative to similar whites than black women in the rest of the country. Not only did occupational opportunities begin to open up and relative wages begin to increase later in the South than in the rest of the country, but they also continue to lag behind. Black women at all ages earn less relative to similarly qualified white women in the South than they do in the rest of the country. Southern black women earn 87 percent as much per hour as comparable southern white women, whereas black women outside of the South earn 96 percent as much as comparable white women. Southern black women earn relatively less because they are underrepresented in middle- and high-status occupations compared with their white counterparts. In particular, given their measured qualifications, black women are considerably under-represented among southern clerical workers. Twenty-three percent of southern black women are clerical workers compared with 36 percent of comparable southern white women.(15) The research in this report suggests that this may be due partially to a North-South differential in the relative quality of education received by black women at middle and higher education levels.(16) Highly educated black women have always fared better relative to similarly qualified white women than their less educated counterparts.(17) In the past, this may have been due to the demand for black school teachers to staff black schools in segregated school systems. This explanation is less likely to account for educated black women's relative success in today's labor market. Another possibility is that highly educated white women themselves face greater discrimination than their less educated counterparts. If this is the case, highly educated black women may appear to be doing better when in fact highly educated white women are doing worse. Labor Market Discrimination and Trends in Wages and Occupations Labor market discrimination against black women exists if employers, co-workers, or customers treat black women differently from white women with identical labor market skills solely on the basis of their race. Thus, if a black woman is paid less, promoted less quickly, denied access to the same job or occupation, or avoided or harassed more by her co-workers than an identical white woman simply because she is black, then she has suffered from labor market discrimination. A central concern of this report is to ascertain the extent to which labor market discrimination against black women has lowered their relative wages and limited their occupational opportunities, both now and in the past. The extent to which black women earn less and are in different occupations than white women with the same characteristics measures the possible effect of labor market discrimination on black women's wages and occupations. This report has found that differences in the measured characteristics of black and white women were able to account for only a small portion of the differences in their wages and occupations. Although differences in unmeasured characteristics, such as the quality of education undoubtedly contributed to the gap in black and white women's wages and occupations, this result suggests that labor market discrimination has played an important part in depressing the wages and occupational status of black women through-out the period from 1940 to the present. Similarly, the report's finding that southern black women had and continue to have lower wages and occupational attainment relative to comparable white women than black women in the rest of the country supports the view that labor market discrimination against black women has been and may continue to be worse in the South than elsewhere. This report has also found that very little of the increase in black women's relative wages and occupational attainment after 1940 can be accounted for by changes in their characteristics. This result suggests that diminishing labor market discrimination, especially declining occupational discrimination, was a major factor responsible for the economic progress experienced by black women over the period. The evidence concerning the present day is less conclusive, because contemporary differences in the wages, occupations. and measured characteristics of black and white women are small by historical standards. There remain, however, portions of the wage and occupational gaps that cannot be explained by differences in the measured characteristics of black and white women, suggesting that current racial discrimination in the labor market may continue to reduce black women's wages and occupational attainment today. This report has focused on uncovering the effects of current racial discrimination on black women's economic status. It should be remembered that, like all women, black women are also subject to gender discrimination, which also lowers their economic status. Furthermore, in addition to current discrimination, the legacy of past discrimination also limits black women's economic status. Whereas, as discussed above, past discrimination certainly continues to restrict opportunities for older black women, younger black women as well may suffer from its legacy. Since social and economic status are generally influenced by upbringing, hardships visited upon their parents and upon the black community by discrimination in the past continue to hinder the progress of young black women today. Employment and Unemployment Patterns This report has found that black and white women have very different labor force participation and unemployment patterns. Regardless of marital status or age. black women have much higher unemployment rates than their white counterparts. Overall, 11 percent of black women who are ln the labor force are unemployed, compared with 4 percent of white women. The black-white unemployment rate differential is greatest for women under 24 years old (24 percent for blacks versus 7 percent for whites) and for unmarried women (14 percent for blacks versus 5 percent for whites).(18) Although black and white women's average labor force participation rates are very close (68 percent and 69 percent, respectively), they have very different labor force participation patterns by marital status and by age. Married black women have much higher labor force participation rates than their white counterparts (73 percent versus 64 percent). Unmarried black women, on the other hand, have much lower labor force participation rates than their white counterparts (56 percent versus 73 percent). Moreover, among whites, married women are less likely than unmarried women to participate in the labor force, but among blacks. the reverse is true. Although older black and white women have similar labor force participation rates, black women under 24 years old are much less likely to participate in the labor force than their white counterparts: 59 percent versus 72 percent. Young black women's lower labor force participation rate combined with their higher unemployment rate means that only 44 percent of black women 24 years old and younger work, compared with 66 percent of white women. To determine whether black-white differences in unemployment and labor force participation rates can be accounted for by racial differences in background characteristics, this report controlled for differences in age, education, presence and age of children, and, for married women, husband's income. Controlling for these characteristics only slightly narrowed the black-white unemployment gap: the report found that black women had substantially higher unemployment rates than white women with the same characteristics. On the other hand, racial differences In characteristics, especially education, presence and age of children, and percentage never-married (as opposed to widowed, separated or divorced), accounted for almost three-quarters of the gap between the labor force participation rates of black and white unmarried women. For married women, however, none of the gap in labor force participation rates could be accounted for by characteristics. The result that the black-white unemployment gap persists after differences in characteristics are controlled for points to the possibility that black women's higher unemployment rates might be caused by employment discrimination rather than by legitimate differences between black and white women. Further, labor market discrimination might lower black women's hours of work indirectly by discouraging their labor force participation. To the extent that labor market discrimination depresses black women's wages and decreases their likelihood of working in the occupation of their choice or their chances of finding a job at all, some black women may respond by choosing not to enter the labor market. Labor Market Discrimination and Black Women's Economic Status The evidence presented in this report suggests that labor market discrimination, although it has abated considerably since 1940, may continue to have adverse effects on black women's economic status today. Wage and occupational discrimination in the labor market may lower black women's average hourly wage in comparison to white women's by as much as 10 percent. Labor market discrimination may also reduce the hours worked by black women: directly by making it difficult for black women to find work and indirectly by reducing their incentive to enter the work force. In addition to current discrimination, past discrimination continues to have a negative effect on black women's economic status. Other factors, such as differences in the family situations of black and white women and the lower earnings of black men (themselves partially the result of labor market discrimination) also have important effects on black women's economic status. Discrimination is reprehensible whenever it occurs. However, since black families are heavily dependent on the earnings of black women, any discrimination experienced by black women can be especially burdensome. Recommendations This report investigates the effect of racial discrimination in the labor market on black women's economic status. A comprehensive assessment of the effect of all forms of discrimination on black women's economic status requires further research. An especially important topic for future research is the effect of gender discrimination on black women's economic status. Also, to complement the statistical analysis contained in this report, new, more refined data sources and research methodologies need to be developed. The findings with respect to black women's economic status and discrimination against black women contained in this report are based upon statistical analysis of data on individuals taken from the 1940-80 Censuses of Population, the 1970-88 March Current Population Surveys (CPS), and the 1984 Panel of the Survey of Income and Program Participation (SIPP). This report has encountered serious inconsistencies among these three data sets that make it difficult to arrive at firm conclusions about the precise level of black women's wages compared to white women's. The census data suggest that black women earn on a par with white women, whereas the CPS and SIPP data suggest that they earn roughly 10 percent less than white women. Research needs to be done to understand the reasons for these inconsistencies and to ascertain black women's relative wage levels with more precision. Perhaps a more serious problem encountered in this report is the difficulty in obtaining accurate measures of women's labor market skills from existing data sources. Accurate measures of women's skills are essential to determining conclusively how much of the wage gap is due to labor market discrimination. One important indicator of women's skills is previous work experience. Of the data sources used in the report, only the SIPP data contain information on women's previous work experience. Because of their historically higher labor force participation rates, black women generally have more work experience than their white counterparts. Using the SIPP data, this report finds that when women's work experience is included in the wage analysis for the 1980s, the unexplained gap in wages, which is often interpreted as encompassing the effects of racial discrimination, increases. When work experience is omitted, black women appear to earn 95 percent as much per hour as comparable white women, and when it is included, they are found to earn 91 percent as much. This result indicates that it is important to include accurate measures of work experience in wage analyses to obtain true estimates of discrimination against black women. Another important dimension of skill omit- ted from all three data sources used in this report is the quality of education. If black women obtain, on average, an inferior education, then measures of educational attainment based on the years of schooling women have completed may overstate black women's true educational achievement relative to white women's. Data sources that do provide information on educational quality would enhance future research into sources of the black-white wage gap. Statistical studies on large national data sets are extremely useful in providing comprehensive information about the likely effects of discrimination and pinpointing problem areas. They have inherent limitations, however. These studies cannot reach definitive conclusions about the existence and extent of labor market discrimination. Furthermore, they yield only modest insight into the nature of labor market discrimination, in particular, the mechanisms through which it operates. Future research can complement statistical studies such as the one in this report by exploring alternative avenues of inquiry. Other data sources may be able to provide more insight into the nature of labor market discrimination. For instance, the Equal Employment Opportunity Commission and State equal opportunity commissions gather information on specific instances of discrimination when complaints are filed. To improve our understanding of the nature of discrimination, it may be fruitful to assemble and analyze information in their files. Another possible avenue for future research is to set up "experiments" to test for employment discrimination by sending black and white women to apply for jobs and monitoring employers' responses. Such experiments would allow researchers to control fully for skill differences by choosing black and white women with very similar skills. They would also have the advantage of providing additional insight into the mechanisms through which labor market discrimination operates. For instance, do employers refuse to interview black applicants? At what points in the hiring process are black women treated differently from their white counterparts? This type of question could be answered by careful monitoring of employers' responses to the black and white job applicants. Experiments might also be useful in antidiscrimination enforcement efforts. By sending carefully matched black and white women to seek jobs, enforcement agencies could obtain direct evidence of illegal discrimination.(19) Although experiments can provide much new information about labor market discrimination, they, too, are limited in their ability to capture fully all aspects of employment discrimination. Although experiments can often be set up to detect discrimination in hiring, it will seldom be practical to set up experiments to detect discrimination in promotions. Furthermore, it is unlikely that experimental evidence can be used to detect hiring discrimination for complex jobs, such as professional jobs and high-level management jobs, that require considerable specialized training and personal contacts. Yet, because of the subjective nature of promotion decisions and hiring decisions for high-level jobs, it is possible that the greatest effect of labor market discrimination in the 1980s is precisely in these areas. Indeed, blacks, women, and other minorities often cite an invisible "glass ceiling" that prevents them from reaching the top of the job ladder.(20) An avenue of research that allows researchers to learn more about discrimination in hiring into top-level jobs and discrimination in promotions is to conduct specialized surveys of individuals or case studies of firms or industries. Specialized surveys could allow researchers to follow the careers of similarly qualified individuals over time and to obtain specific information about their qualifications, their job applications. the times when they were up for promotion, and so on. Case studies could allow researchers to look closely at employers' decision making processes when choosing whom to hire or whom to promote. Both of these types of studies would add significantly to the current understanding of labor market discrimination. Several of the findings in this report would benefit from further scrutiny using one or more of the approaches outlined above. First, the report finds that black women experience higher unemployment rates than comparable white women, but it does not determine whether the black-white unemployment gap among women is caused by racial discrimination in the labor market. Given the importance of working for black women's economic well-being, further study of this issue seems warranted. A more detailed statistical study of the sources of the black-white unemployment gap, taking into account such factors as labor market turnover and residential location in comparison to the location of jobs could be complemented with the experimental approach--sending black and white women with similar qualifications to apply for jobs--to determine whether black women are denied jobs with greater frequency. In addition, specialized surveys could be undertaken to address questions related to whether workers become discouraged when they face high unemployment rates as well as to determine whether there are differences in the job search strategies of black and white women. Similarly, the report's finding that black women are underrepresented among clerical workers in the South could be further investigated using all of the above approaches. Specialized surveys could provide better evidence pertaining to racial differences in the quality of education in the South. Experiments could be set up to determine whether equally qualified black and white women had different chances of finding clerical work. The alternative employment opportunities for black and white women could also be compared. This report provides evidence supporting the view that racial discrimination continues to affect black women in today's labor market and pinpoints several problem areas. On the whole, the evidence presented in this report suggests that the main effect of labor market discrimination on the basis of race today is to limit black women's occupational opportunities, particularly in management and sales jobs. In the South, black women appear also to have substantially fewer employment opportunities in clerical occupations than comparable white women. Thus, a major problem facing black women today appears to be discrimination in hiring, referrals and promotions. Since discrimination in hiring, referrals and promotions can be extremely subtle, identifying and combatting employment discrimination in these areas is inherently difficult. Often the victims of discrimination may not even be aware that it has occurred. Consequently, new and aggressive enforcement methods may be needed to eradicate discrimination against black women. One such method, audits of firms and employment agencies, might be useful in antidiscrimination enforcement efforts: By sending carefully matched individuals of different races and genders to apply for jobs, enforcement agencies could obtain direct evidence of illegal discrimination in hiring or referrals. Such direct evidence would not only be useful in prosecuting discrimination cases, but could also provide a valuable informational basis for guiding antidiscrimination enforcement policy. As an example, the New York City Human Rights Commission is currently using evidence gathered by Commission employees posing as job applicants in prosecuting four employment agencies for discriminating against blacks, Hispanics, women, and the elderly.(21) Other agencies in charge of enforcing equal opportunity legislation, such as the Equal Employment Opportunity Commission, should consider using audits to ferret out discrimination in employment and should search actively for other innovative means of enforcing anti-discrimination laws. ENDNOTES (1) See table 9.2. (2) See U.S Commission on Civil Rights, The Economic Progress of Black Men (1986) for a study of labor market discrimination against black men. (3) See table 8.2. (4) The 1980 census suggests a figure of 100 percent, whereas the 1980-87 March Current Population Surveys and the Survey of Income and Program Participation suggest figures closer to 90 percent. Appendix A discusses the discrepancies between the census and the Current Population Survey data. (5) Other factors considered included age, region of residence, urban or rural residence, and part time or full time work status. (6) The term "comparable white women" refers to white women with the same characteristics as the average black woman. (7) When racial differences in occupation and industry were not considered, racial differences in education by themselves accounted for black women earning 19 percent less than white woman in 1940, or roughly one-third of the black-white wage differential. For 1960, differences in education accounted for black women earning 10 percent less than white women, or roughly one-quarter of the differential. In 1980, differences in education accounted for black women earnings 10 percent less, roughly two-thirds of the gap. see table 6.2. (8) In 1940, when black women earned 63 percent less than white women, occupational differences between black and white women by themselves accounted for black women earning 43 percent less, or roughly two-thirds of the differential. See table 6.4. (9) In 1980 black women earned roughly 6 percent less than white women, and occupational differences accounted for their earning 4 percent less or roughly two-thirds of the black-white differential. See table 6.4. (10) See table 6.9. (11) Black women's increased relative education accounted for roughly 13 percent of their increased relative wage. Changes in regional distribution and urban-rural residence accounted for 6 and 5 percent of black women's increased relative wage, respectively. See table 6.8. (12) See table 5.14. (13) See app. C. (14) Twenty-nine percent of black women are clerical workers, compared with 36 percent of comparable white women. As discussed below, black women's under-representation among clerical workers is particularly prominent in the South. (15) See app. B table B.10. Outside of the South 35 percent of black women are clerical workers, compared with 37 percent of comparable white women. See app. B., table B.11. (16) See app. C for an analysis of educational achievement by region for black and white women. (17) See table 6.6. (18) See table 9.1. (19) Of course, such experiments would not necessarily be limited to black and white women. White men, black men, and men and women from other minority groups could also be used in the experiments. Discrimination in other areas besides the labor market could also be addressed with experiments. For instance, housing discrimination and discrimination in lending might be easily detected from experimental evidence. (20) Statistical evidence of discrimination at the top was found in U.S> Commission on Civil Rights, The Economic Status of Americans of Asian Descent: An Exploratory Investigation (1988). (21) "New York Sues 4 Work Firms in Bias Case," New York Times, Sept. 29, 1989. Bibliography Aigner, Dennis J. and Glen C. Cain. 1977. "Statistical Theories of Discrimination in the Labor Market." Industrial and Labor Relations Review 30:175-89. Allen, Walter R. 1979. "Family Roles, Occupational Statuses, and Achievement Orientations Among Black Women in the United States." Journal of Women in Culture and Society 4:670-86. --------. 1981. "The Social and Economic Statuses of Black Women in the United Sates." Phylon 42:26-40. Arrow, Kenneth J. 1973. "The Theory of Discrimination." In Discrimination in Labor Markets, ed. Orley Ashenfelter and Albert Rees, pp. 3-33. Princeton, N.J.: Princeton University Press. --------. 1972a. "Models of Job Discrimination." In Racial Discrimination in Economic Life, ed. Anthony H. Pascal, pp. 83-102. Lexington, Mass.: Lexington Books, D.D. Heath and Co. --------. 1972b. "Some Mathematical Models of Race in the Labor Market." In Racial Discrimination in Economic Life, ed. Anthony H. Pascal, pp. 187-204. Lexington, Mass.: Lexington Books, D.D. Heath and Co. Becker, Gary S. 1957. The Economics of Discrimination. Chicago: University of Chicago Press. Bell, Duran 1974. "Why Participation Rates of Black Wives Differ." Journal of Human Resources 9:465-79. Benus, Jacob M. 1974. "Income Instability." In Five Thousand American Families: Patterns of Economic Progress, pp. 277-303. Ann Arbor, MI: Survey Research Center. Bergmann, Barbara. 1974. "Occupational Segregation, Wages and Profits When Employers Discriminate by Race or Sex." Eastern Economic Journal 1(2):103-10. Bowles, Eva D. 1923. "Opportunities for the Educated Colored Woman." Opportunity 1:8-10. Bowen, William G. and T. Aldrich Finegan. 1969. The Economics of Labor Force Participation, pp. 88-269. Princeton, NJ: Princeton University Press. Bulow, Jeremy I. and Lawrence H. Summers. 1986. "A Theory of Dual Labor Markets with Application to Industrial Policy, Discrimination, and Keynesian Unemployment." Journal of Labor Economics 4(3, part 1):376-414. Cain, Glen G. 1966. Married Women in the Labor Force: an Economic Analysis. Chicago: University of Chicago Press. Calhoun, Charles A. and Thomas J. Espenshade. 1988. "Childbearing and Wives' Foregone Earnings." Population Studies 42:5-37. Carliner, Geoffrey 1981. "Female Labor Force Participation Rates for Nine Ethnic Groups." Journal of Human Resources 16:286-93. Carlson, Leornard A. and Caroline Swartz. 1988. "The Earnings of Ethnic Minorities, 1959-1979." Industrial and Labor Relations Review. 41:530-552. Chafe, William H. 1972. The American Woman: Her Changing Social Economic, and Political Roles, 1920-1970. New York: Oxford University Press. Corcoran, Mary and Greg J. Duncan. 1979. "Work History, Labor Force Attachment and Earnings Differences Between Races and Sexes." Journal of Human Resources 1:1-20. Cunningham, James S. and Nadja Zalokar. 1988. "Racial Differences in the Labor Force Participation of Wives and the Variability of Other Family Income: Evidence from the SIPP." Paper presented at the annual meetings of the American Economic Association in New York. Darity, William A. Jr. 1982. "The Human Capital Approach to Black-White Earnlngs Inequality: Some Unsettled Questions." Journal of Human Resources 17(1):72-93. ------- and Samuel L. Myers, Jr. 1984. "Does Welfare Dependency Cause Female Headship? The Case of the Black Family." Journal of Marrieg and the Family 46(4):- 765-79. Davis, Lenwood G. 1972. Black Women in the Cities, 1872-1972: a Bibliography of Published Works on the Life and Achievements of Black Women in Cities in the United States. No. 336. Monticello, IL: Council of Planning Librarians. -------. 1975. The Black Woman in American Society: a Selected Annotated Bibliography. Boston: G.K. Hall & Co. Douglas, Priscilla H. 1980. "Black Working Women: Factors Affecting Labor Market Experience." Wellesley, MA: Center For Research on Women. Ehrenberg, Ronald G. 1980. "The Demographic Structure of Unemployment Rates and Labor Market Transition Probabilities." Research in Labor Economics 3:241-93. Farley, Reynolds. 1988. "After the Starting Line: Blacks and Women in an Uphill Race." Demography 25(4):447-95. and Walter R Allen. 1987. The Color Line and the Quality of Life in America. New York: Russell Sage Foundation. Freeman, Richard B. 1973. "Decline of Labor Market Discrimination and Economic Analysis." American Economic Review 63(2):280-86. Freeman, Richard B. 1974. "Labor Market Discrimination: Analysis, Findings and Problems." Chapter 9 in Michael D. Intriligator and David A Kendrick, eds., Froniers of Quantitative Economics. Amsterdam: North-Holland. Giddings, Paula. 1984. When And Where I Enter: The Impact of Black Women on Race and Sex in America New York: William Morrow and Co. Goldin, Claudia. 1977. "Female Labor Force Participation: The Origins of Black and White Differences, 1870 and 1880." Journal of Economic History 37:87-108. Harley, Sharon and Rosalyn Terborg-Penn., eds. 1978. The Afro-American Woman Struggles and Images. Port Washington, NY: National University Publications. Heckman, James J. 1979. "Sample Bias as a Specification Error." Econometrica 47:-153-62. Heckman, James J. and Brook S. Payner. 1989. "Determining the Impact of Federal Antidiscrimination Policy on the Economic Status of Blacks: a Study of South Carolina." American Economic Review 79(1):138-77. Hoffman, Emily P. 1987. "Determinants of Youth's Educational and Occupational Goals: Sex and Race Differences." Economics of Education Review 6:41-48. Hudis, Paula M. 1977. "Commitment to Work and Wages: Earnings Differences of Black and White Women." Sociology of Work and Occupations 4:123-46. Jones, Elizabeth D. 1987. Women, Race and Work: A Study of Transitions to Adulthood. Ph.D. diss., Northwestern University. Jones, Jacqueline. 1985. Labor of Love, Labor of Sorrow. New York: Basic Books. Klotman, Phyllis R and Wilmer H. Baatz. 1978. The Black Family and the Black Woman. New York: Mae Arno Press. Lerner, Gerda, ed. 1973. Black Women in White America: a Documentary History. New York: Vintage Books. Lloyd, Cynthia B., Emily S. Harris, and Curtis Gilroy. 1979. Women in the Labor Market. New York: Columbia University Press. Mae, Verta. 1972. Thursdays and Every Other Sunday Off. Garden City, NY: Doubleday & Co. Malveaux, Julianne. 1984. "Theoretical Explanations of the Persistence of Racial Unemployment Differentials." pp. 91-118 in William Darity, Jr., ed., Labor Economics: Modern Views Boston: Kluwer-Nijhoff. Margo, Robert A. 1986a. "Race, Educational Attainment, and the 1940 Census." Journal of Economic History 46:189-98. -------. 1986b. "Race and School Attendance in the American South: Evidence from the 1900 Census Sample." Mimeograph. Philadelphia: University of Pennsylvania. Mincer, Jacob and Haim Ofek. 1982. "Interrupted Work Careers: Depreciation and Restoration of Human Capital." Journal of Human Resources 7(1):3-24. Mincer, Jacob and Solomon Polachek. 1974. "Family Investments in Human Capital and the Earnings of Women." Journal of Political Economy 82(2):S76-108. Patterson, George. 1985. "Quality and Comparability of Personal Income Data from Surveys and the Decennial Census." Paper presented at the Plenary Session of the Joint Advisory Committee Meeting, April 25-26, 1985 in Rosslyn, Virginia. Phelps, Edmund S. 1972. "The Statistical Theory of Racism and Sexism." American Economic Review, 62:659-61. Reimers, Cordelia W. 1985. "A Comparative Analysis of the Wage of Hispanics, Blacks and Non-Hispanic Whites." In Hispanics in the U.S. Economy. Eds. G.J. Borjas and M. Tienda. 27-76. Orlando, FL: Academic Press. Rowan, Richard L. 1970. The Negro in the Textile Industry. Report No. 20 in The Racial Policies of American Industry. Philadelphia: Industrial Research Unit, Wharton School of Finance and Commerce, University of Pennsylvania. Sandell, Steven and David Shapiro. 1978. "An Exchange: The Theory of Human Capital and the Earnings of Women." Journal of Human Resources 13(1):103-117. Simms, Margaret C. and Julianne M. Malveaux. 1987. Slipping Through The Cracks: the Status of Black Women. New Brunswick, NJ: Transaction Books. Sims, Janet L. 1979. Black Women in the Employment Sector. Public Administration Series, No. P-243. Montecello, IL: Vance Bibliographies. Smith, James P. 1979. "The Distribution of Family Earnings." Journal of Political Economy S163-S192. -------. 1979. "The Convergence to Racial Equality in Women's Wages." In Women in the Labor Market. Eds. C.B. Lloyd, E.S. Andrews, and C.L. Gilroy. 173-215. New York: Columbia Universlty Press. -------. 1984. "Race and Human Capital." American Economic Review 77:685- 98. -------. and Finis Welch. 1978. "Race Differences in Earnings: a Survey and New Evidence." No. R-2295-NSF. Santa Monica, CA: Rand Corporation. Stolzenberg, Ross M. 1975. "Education, Occupation, and Wage Differences Between White and Black Men." American Journal of Sociology 81:299-323. Stone, Louise. 1966. "What It's Like to Be a Colored Woman." Washington Post, November 13, 1966, p.43. Sweet, James A. 1973. Women in The Labor Force. New York: Seminar Press. "The Changing Status of Negro Women Workers." 1964. Monthly Labor Review, 87:- 671-73. Tienda, Marta, Shelley A. Smith and Vilma Ortiz. 1987. "Industrial Restructuring, Gender Segregation and Sex Differences in Earnings." American Sociological Review 52: 195-2 10. Turner, Ralph H. 1951. "The Nonwhite Female in the Labor Force." American Joumal of Sociology 56:438-47. U.S. Commission on Civil Rights, Office of Programs, Policy, and Research. 1990. "Technical Appendix." Available upon request from the Office of Programs, Policy, and Research, Room 700, U.S. Commission on Civil Rights, 1121 Vermont Avenue NW, Washington, DC 20425. U.S. Department of Commerce, Bureau of the Census. 1986. Suruey of Income and Program Participation: 1984 Panel, Waves 1-9. Washington, D.C. U.S. Department of Commerce, Bureau of the Census. 1984. Census of Population, 1940: Public Use Microdata Sample. Washington, D.C. U.S. Department of Commerce, Bureau of the Census. 1984. Census of Population, 1950: Public Use Microdata Sample. Washington, D.C. U.S. Department of Commerce, Bureau of the Census. 1983. Census of Population, 1980: Public Use Microdata Sample. Washington, D.C. U.S. Department of Commerce, Bureau of the Census. 1972. Census of Population, 1970: Public Use Microdata Sample. Washington, D.C. U.S. Department of Commerce, Bureau of the Census. 1963. Census of Population, 1960: Public Use Microdata Sample. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1988. Current Population Survey, March 1988. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1987. Current Population Survey, March 1987. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1986. Current Population Survey, March 1986. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1985. Current Population Survey, March 1985. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1983. Current Population Survey, March 1983. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1980. Current Population Survey, March 1980. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1971. Current Population Survey. March 1971. Washington, D.C. U.S. Department of Commerce, Bureau of the Census for the Bureau of Labor Statistics. 1970. Current Population Suruey, March 1970. Washington, D.C. Wallace, Phyllis A., ed. 1980. Black Women in the Labor Force. Cambridge, MA: MIT Press. Wilkie, Jane R 1985. The Decline of Oc- cupational Segregation Between Black and White Women." In Research in Race and Ethnic Relations, Vol. 4. Eds. C.B. Marrett and C. Leggon, pp. 67-89. Greenwich, CT: JAI Press. Williams, Walter E. 1982. The State Against Blacks. New York: New Press. Woodman, Harold. 1977. "Comment." Journal of Economic History 37:109-12.