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.
I~ Occupational Desegregation in C' CETA Programs LINDA I. WAITE and SUE E. BERRYMAN This paper examines the occupational dis- tributions in the Comprehensive Employ- ment and Training Act (CETA) and the wage implications of these distributions for men and women of different racial and ethnic origins. Our data come from two separate projects on CETA, both conducted for the National Commission on Employment Pol- icy. One (Berryman et al., 1981) assessed the nature and equity of men's and; women's experiences in CETA, a substantial part of the study being devoted to CETA's occu- pational desegregation record for women. The second (Berryman and Waite, 1982) as- sessed ethnic and racial differences in CETA experiences, focusing on whites, blacks, anal Hispanics and on Hispanic subgroups. CETA's occupational desegregation rec- ord for women is important for several rea- sons. First, one of CETA's legislated pur- poses is to improve the economic prospects of its clients. As we know, substantially more female than male occupations pay poverty- level wages (Sawhill, 19761. Persistent oc- cupational segregation parallels the persist- ent male-female wage differential, and dif- ferences in male and female occupational distributions account for over a quarter of 292 the wage differential (Chiswick et al., 19741. Even when labor force attachment is con- trolled, women also have much flatter life- time earnings profiles than do men (Sawhill, 1973~. Theoretical arguments (Wolf and Ro- senfeld, 1978) and fragmentary evidence (Barrett, 1979) implicate occupational seg- regation in these profile differences. Male but not female occupations seem associated with career paths that carry wage advance- ment with experience. Second, poverty in the United States is becoming increasingly female poverty, pri- marily as the result of the rising number of female-headed households and the relation- ship between households of this kind and poverty. Thus, from the economic per- spective, the issue of occupations and wages for women is not transitory. ~ From 1969 to 1979 the percentage offemale-headed households of all races increased by a third. For whites and Hispanics the increase was about 25 percent; for blacks, over 40 percent. Although the chances that a household of this kind was poor declined slightly over the decade, in 1979 they were still very high: 30 per- cent for all races and almost 50 percent for black female- headed households (Bureau of the Census, 1981~.
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS 293 Third, CETA has represented a major federal lever for affecting occupational de- segregation for women and women's wages. From FY 1974 to FY 1980, 19 million in- dividuals entered CETA in job training or employment capacities, somewhat fewer than half of these being women. Thus, over time CETA has had the potential for affecting the occupational preferences and skills of large numbers of women. Finally, CETA flows from early federal manpower programs of the 1960s and can be expected to affect future federal training and employment programs. Thus, even if CETA is virtually dismantled under the Reagan administration, its occupational de- segregation record for women is of more than historic interest. As we show later, women's occupational options in CETA are affected by how CETA is structured and by how men and women are funneled through this structure. Our experience with CETA has implications for designing future pro- grams that would increase women's expo- sure to occupations currently held mostly by men. The paper has five sections. The first briefly describes CETA's legal structure its ti- tles, their legislated purposes, and eligibil- ity rules. The second describes the data base used in the two studies that underlie this paper. The third shows how the CETA title under which individuals enter CETA and their CETA activity (e.g., work experience) affect their occupational options. The fourth documents CETA's occupational desegre- gation record for white, black, and Hispanic women; and the final section shows the wage consequences of women's occupational dis- tributions in CETA. DESCRIPTION OF CETA TITLES AND ELIGIBILITY REQUIREMENTS For reasons of simplicity and of data re- strictions, we deal only with CETA Titles I, II, and VI.2 The major services available under these titles were basic skills, job train- ing, and jobs, although, as we describe later, not all services are available in all titles. For example, basic skills and job training are essentially restricted to Title I. The pur- poses of the jobs also vary by title. Most Title I jobs, called work experience, are in- come transfer jobs3 that are not intended as a bridge to unsubsidized employment. Jobs in Titles II and VI, known as public service employment (PSE) jobs, are expected to lead to unsubsidized employment, although the economic environments in which these jobs are offered presumably vary. Title II jobs are available in areas with high, structural unemployment; Title VI jobs, in areas with short-term, cyclical unemployment.4 The eligibility by title varied, although, as we discuss below, titles overlapped in their eligibility requirements. All of the ti- tIes had eligibility criteria of economic dis- advantage, underemployment, or unem- ployment. For Title I, eligibility was restricted to those economically disadvan- taged or unemployed or underemployed.5 2 These are the title numbers before the 1978 reau- thorization of CETA; they correspond to the postreau- thorization numbers of JIB, IID, and VI. This paper does not discuss Title III because most slots in this title (Title IIIA or the Summer Youth Program) are jobs of short duration, intended as a mechanism of income transfer, and without a training component. 3 By "income transfer jobs" we mean jobs used pri- marily as a means of allocating money to people, not as bridges to private or public sector jobs unsubsidized by CETA. 4 Title II was targeted on regions with lingering un- employment. Title VI was designed to reduce the pre- sumably short-term unemployment associated with the recession of the mid-1970s. However, as Mirengoff and Rindler (1978) observe, the unemployment rate used to define an area's eligibility for Title II was surpassed in most places by that used to define an area's eligibility for Title VI funds. Thus. de facto the distinction be- tween the two titles was eliminated. 5 To receive one of the small number of PSE jobs in Title I the individual had to be unemployed or under- employed.
2 ~ LINDA [. WAITE AND SUE E. BERRYMAN For Title II, individuals had to reside in areas of substantial unemployment. They also had to be unemployed for at least 30 days prior to application or underemployed. Before January 1977 indivicluals were el- igible for Title VI if they had been unem- ployed for 30 days, or if they resided in an area with excessive unemployment and had been underemployed or unemployed for at least 15 days. After January 1977 the eligi- bility rules became more complicated. However, in general, individuals could en- ter if they were (1) unemployed or under- employed; or (2) a member of an economi- cally disadvantaged family and either a member of an Aid to Families With De- pendent Children (AFDC) family, or un- employed and an unemployment insurance recipient, or unemployed and ineligible for unemployment insurance, or unemployed and an insurance exhaustee. DATA Continuous Longitudinal Manpower Survey (:CLMS) Both studies on which this paper is based used Continuous Longituclinal Manpower Survey (CLMS) data. The Bureau of the Census has conducted the CLMS quarterly since January 1975, sampling respondents from the previous quarter's new enrollees in CETA. Respondents are sampled from four CETA functional activities: public serv- ice employment, employability develop- ment, direct referrals, 6 and youth work ex- perience (including summer programs). The CLMS has two main objectives. First, it is designed to obtain characteristics of the CETA participants and the services they re- ceived, thus providing data not available from 6 In a direct referral, CETA refers the individual to a job vacancy. The individual does not receive any other CETA services and does not necessarily get the job to which he or she is referred. the usual sources, the prime sponsors re- porting system. Second, the CLMS is in- tended to measure the effect of CETA pro- grams on participants, including earnings and labor force status.8 The CLMS comprises an initial intake in- terview, an activity record, and several other interviews during and after the CETA en- rolIment. In the initial interview, the CLMS determines what CETA service the enrollee received (e.g., public service employment) and, if the service was a job or job training, the enrollee's occupation and wages. The CLMS also obtains information on the en- rollee's attitudes toward manpower pro- grams and services received, what service and occupation the enrollee wanted from CETA, his or her trade or vocational train- ing before entering CETA, veteran status, marital status, number of dependents, fam- ily composition, receipt of government transfer payments (food stamps, subsidized housing, AFDC, Supplemental Security In- come, unemployment benefits, and other public assistance), the enrollee's employ- ment/schooling history in the previous year, wages or salary in the last year, and personal and family income by source. The CLMS contains information on the highest grade or year of regular school the enrollee at- tended, whether that grade had been com- 7 The federal government administers decentralized CETA programs through administrative units called prime sponsors. Federal funds for these programs are allocated to the prime sponsors. State, county, or local governments can be prime sponsors if they govern a minimum population of 100,000. State governments tend to become prime sponsors for governmental units within the state that do not meet the minimum pop- ulation requirement. 8 The CLMSsponsored by the Employment and Training Administrationsamples mainly decentral- ized CETA programs, i.e., programs operated by CETA prime sponsors. Thus, special-purpose programs such as the Job Corps (Title IV, reauthorized as Title IVB), Young Adult Conservation Corps (Title VIII), and sev- eral Title III (reauthorized as various Title IV) programs are not included in the CLMS file.
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS _ 295 pleted by the time of CETA enrollment, and whether the enrollee had a high school equivalency certificate or General Educa- tional Development (GED) certificate. Our analysis relies especially on detailed data on the enrollee's ethnic origin or descent and on whether the enrollee was limited in the amount or type of work he or she could do because of problems in speaking English. We use all of this information in various sections of our analysis, either as dependent or as independent variables. In our analyses we use only data from the initial questionnaire and activity record, since our purpose is to assess the services pro- vided within CETA ant! not to assess the impact of CETA services on later outcomes. We included all CETA enrollees surveyed by the CLMS during the period October 1975 through September 1978 in order to (1) maximize the period covered by our analyses and (2) maximize the number of cases available for analysis.9 In each quarter the CLMS sampled between 3,500 and 4,000 CETA enrollees and completed initial in- terviews with 3,300 to 3,600. To have suf- ficient numbers of observations for race/eth- nic groups by sex we pooled information for all quarters in the October 1975 to Septem- ber 1978 or March 1979 time period (Bar- rett, 19791. Pooling observations across time periods provides large sample sizes that al- low us considerable flexibility in the types of analyses we do and that allow us to dis- 9 We begin with October 1975 because the CLMS did not record CETA title until the second quarter of FY 1976 (October 1975~. The sample for the multivar- iate analysis ends with March 1979 because CETA was reauthorized in October 1978 and regulations govern- ing the revised act were released to prime sponsors in April 1979. Since those enrolled in CETA in the third and fourth quarters of 1979 entered under revised guidelines, the data for these quarters are not com- pletely comparable with early data, and we eliminated them to ensure comparability. The sample for the cross- tabular analysis ends with September 1979 because data to this date only were available at the time this analysis was done (Berryman et al., 1981~. aggregate the sample by sex and race/eth- nicity. For the October 1975 to March 1979 period, the CLMS contains approximately 42,000 initial interviews. Analytic Strategy We assessed the impact of race and eth- nicity on enrollees' experience in CETA in two ways.~° First, we estimated a general linear mode} of each CETA outcome sepa- rately for men and women in which we con- trolled for all characteristics of the enrollee and the enrollment that were relevant for CETA assignment. i~ This mode! included a series of dummy variables for race/ethnicity: white, black, and Hispanic. Second, we per- formed an analysis of covariance for each CETA outcome in which we tested for dif- ference between race/ethnic groups in the slope coefficients in the model. CETA AS A SYSTEM OF OPPORTUNITIES We can think of CETA as a system for distributing opportunities of several kinds: (1) participation in CETA; (2) a CETA serv- ice or activity- basic education, job train- ing in a classroom setting, on-thejob train- ing, work experience, and public service employment; (3) an occupation for those in jobs or job training; and (4) a CETA wage for those in jobs or job training. Since this paper focuses on CETA's contribution to oc- cupational desegregation for women, CETA occupations are the resource of primary con- cern. However, to interpret the data on oc- ~° We follow census definitions; persons of Hispanic origin may be of either race. We divide enrollees into whites (non-Hispanic), blacks (non-Hispanic), and His- panics of both races. We omit those of other races who are not Hispanic. ii These included age, marital status, poverty status, labor force experience, educational attainment, desired CETA services, and problems with the English lan- guage all at the time of enrollment, plus, for males, veteran status.
296 LINDA I. WAITE AND SUE E. BERRYMAN cupations it is important to understand the process by which a CETA participant is as- signed an occupation, either in the form of a job or job training. An individual enters CETA under a title and a CETA activity that is authorized for that title. If the activity is job training or a job, the individual is assigned to an occu- pation and receives a wage in connection with it. Eligibility rules determine if an in- dividual can enter CETA, and under what title he or she may enter. Although these rules vary for different CETA titles, indi- viduals can be eligible for more than one title, giving CETA prime sponsors some dis- cretion in their title assignments. Titles affect CETA service or activity as- signments in that not all CETA services are available in all titles. Titles II and VI consist only of public service employment (PSE) jobs, and almost all of these jobs occur in these two titles. Title I consists primarily of basic education, job training in a classroom setting, on-thejob training (OlT), and work experience activities, and these services oc- cur only in Title I. In sum, Titles II and VI imply a public service job; Title I, a basic education, job training, or work experience activity. If a CETA participant is only eli- gible for Titles II or III, his or her CETA activity is determined. If the participant is only eligible for Title I, his or her activity options are constrained but not determined. As Table 16-1 shows, each CETA service has a different occupational distribution and therefore different occupational assignment probabilities. All of the occupations avail- able in CETA are available in each of the services, but the occupational emphases dif- fer for each CETA service. Relative to the distributions for the other services, cIass- room training has the highest percentage of clerical openings; on-thejob training, the highest percentages of crafts and operatives options; work experience, the highest per- centage of service jobs; and public service employment, the highest percentages of professional/technical and laborer jobs. TABLE 16-1 CETA's FY 1976-FY 1979 Occupational Structure by CETA Activity (percent) CETA's Occupational Structure All CETA Classroom Activities Training Occupational Categorya Professional/technical Managerial/administrative Sales workers Clerical Crafts Operatives Transportation equipment operatives Laborers Service Totals N (000) a These are the 1-digit census occupational categories. They exclude three categories that do not occur in the CETA occupational structure: Farmers and Farm Managers, Farm Laborers and Supervisors, and Private House- hold Workers. b Columns may not add to 100 due to rounding. SOURCE: Table 31, Berryman and Waite (1982), p. 79. OJT 4.9 2.9 3.7 16.3 21.5 28.0 Public Work Service Experience Employment 6.7 0.9 0.9 32.1 6.7 4.2 10.7 2.3 1.0 27.2 12.0 2.7 15.2 21.5 100.0 2,770 6.9 0.4 1.2 38.0 20.3 14.9 1.2 1.2 15.8 100.0 389 3.6 8.6 10.4 100.0 319 15.8 3.6 0.3 23.5 10.3 2.1 2.1 13.8 32.8 100.0 3.3 22.0 19.0 100.0 790 1,272
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS 297 The process by which a CETA participant is assigned an occupation varies across time for any given CETA office and across CETA offices at any given time. Any specific oc- cupational assignment reflects several fac- tors: (1) the participant's title eligibility and the subsequent activity and occupational constraints that are associated with each ti- tie; (2) the participant's activity and occu- pational needs and preferences; (3) the ac- tivity and occupational vacancies available at any given time as a function of the local labor market, the CETA office's efforts to develop particular opportunities, and the amount of federal money then available for different CETA titles; and (4) judgments by CETA assignment personnel about what kinds of occupations are appropriate for what kinds of people. Our analyses show that in FY 1976 to FY 1978, relative to their eligibility, women 18 to 65 years of age were underrepresented in all CETA titles for all three fiscal years except Title I in FY 1978. The underrepre- sentation varied from 64 to 87 percent, de- pending on fiscal year and title. Incus, women did not receive CETA resources, including occupational experiences, at rates commen- surate with their eligibility. The discrepancy between eligibility and participation was greater for Titles II and VI than for Title I. i2 When we looked at how female and male CETA participants distributecl across titles, a logistic regression showed that relative to men with the same placement-relevant characteristics (see note 11), women were more likely to enter CETA under Title I and less likely to enter CETA under Titles II and VI. Thus, they were more likely than t2 Available eligibility estimates for this time period are by sex and by race and other ethnicity separately. Thus, we cannot assess racial and ethnic differences in women's CETA participation, relative to eligibility. We can note that, relative to eligibility, whites are under- represented and blacks are overrepresented in all three titles, and Hispanics are overrepresented in Title I and underrepresented in Titles II and IV. were men to receive basic eclucation, job training, and work experience services and less likely to get public service jobs. Although sex affected title assignment, Berryman and Waite (1982) found few ef- fects anal no important effects of race/ ethnicity on the CETA title under which enrollees enter CETA. Whites of both sexes entered CETA under Titles I and II slightly more often than did blacks or Hispanics with similar characteristics. But these differences never exceeded about 3 percentage points and, although statistically significant, were hardly substantively so. As noted, Title I consists of several CETA services: basic education in a classroom, job training in a classroom setting, OlT, work experience, and a small number of PSE jobs. Again, multivariate analyses showed that race and ethnicity had no or only trivial effects on assignment to CETA services. However, relative to males in Title I, women in this title were placed more frequently in cIass- room training and work experience jobs and less frequently in OIT and PSE jobs. Al- though the percentages declined across fis- cal years, even in FY 1978 a third of all women in CETA were in Title I classroom training. Thus, relative to men's occupational op- tions, women's options were more apt to be those associated with classroom training and work experience. They were less apt to be those associated with OIT and PSE jobs. We would like to use multivariate anal- yses to assess CETA's occupational sex seg- regation for racial and ethnic groups. How- ever, Berryman et al. (1981) did not conduct multivariate analyses of occupational seg- regation by sex and race and ethnicity. Ber- ryman and Waite (1982) conduct multivar- iate analyses separately by sex and by race and ethnicity an :1 have no direct measure of the sex composition of occupations for these groups. The occupational measure used in Berryman and Waite was occupational sta- tus, a measure that does not directly bear on occupational segregation. However, we
ZD9 LINDA I. WAITE AND SUE E. BERRYMAN can use our multivariate results for the ef- fects of race and ethnicity on CETA title, CETA service, occupational status, and CETA wages to draw tentative inferences about these effects on occupational segre- gation in CETA by race and ethnicity. We have already noted that there are few, if any, effects of race and ethnicity on CETA title and service assignments. Our analyses of the impact of race/ethnicity on the oc- cupational status of CETA jobs and job train- ing showed mixed results. As Table 16-2 shows, we found no differences among white, black, and Hispanic males in status of job training, but we found lower occupational status for white ant] black than for Hisuanic females, net of other characteristics. For job status we found lower scores for whites and blacks of both sexes than for Hispanics. As before, the differences tended to be statis- tically significant but substantively unim- portant. The largest coefficient for race/eth- nic groups appeared for black females in occupational status of job training and equaled 5 points on a 100-point scale, the Duncan Socioeconomic Index. Our results for race/ethnic differences in CETA wages, shown in Table 16-3, rein- TABLE 16-2 Effects of Race and Ethnicity on Occupational Status of CETA lob Training and CETA lob Unstandardized Regression Coefficients, Net of Social, Economic, and Demographic Characteristics of the CETA Entranta Males Females Job training White Black Job White - 2.4285b Black 2. 7244b .5086 .5447 _ 3.3100b 3. 9040b _ 3.3197b _ 4. 9859b a These characteristics are listed in note 11 of this paper. b p < .05. This indicates effects that would appear by chance less than 5 times out of every 100 analyses. SOURCE: Tables 42 and 43, Berryman and Waite (1982), pp. 97 and 99. TABLE 16-3 Effects of Race and Ethnicity on Hourly Wage of CETA Job Training and CETA Job Unstandardized Regression Coefficients, Net of Social, Economic, and Demographic Characteristics of the CETA Entranta Males Females Job Training White Black Job White Black .0001 - .0225 _ . o555b .0065 . 0396b . 0149b . 0285b a These characteristics are listed in note 11 of this paper. b p < .05. This indicates effects that would appear by chance less than 5 times out of every 100 analyses. SOURCE: Tables 46 and 47, Berryman and Waite (1982), pp. 105 and 107. forced the conclusions we reached for cupational status. Table 16-2 presents re- sults of the regression of the Duncan Socioeconomic Index of CETA job training or CETA job on assignment-relevant char- acteristics (see note 11) of the individual. Since these models omit the variable for the "Hispanic" race/ethnicity category, the coefficients show the deviation of white and black occupational status from that for His- panics, controlling for the other character- istics of the enrollee. Among males in job training, we fount] no differences in wages, but among males in jobs, black males re- ceived wages 4 percent lower than those of Hispanic and white males with comparable characteristics. For females, we found very small differences on the order of 1 or 2 percent- but those that did exist favored Hispanics. The analyses of covariance allowed us to test the hypothesis that the process which determines CETA occupational status and wages clepencls on race/ethnicity. We found evidence of some rather minor differences. These analyses showed cli~erent effects of the variables in the models for race/ethnic groups on occupational status of CETA job
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS 299 training for males but not for females and for status of CETA jobs for both males and females. But few sizable differences ap- peared in individual coefficients in any of these models. In sum, when we considered males and females separately, we found small or no effects of race/ethnicity on CETA title, ac- tivity, occupational status, or wages. The differences that existed tended to favor His- panics over blacks and whites. These results suggest that, given an appropriate measure of occupational segregation, we would have found that the process of occupational seg- regation dicl not depend on race and eth- nicity. The remainder of this paper focuses on CETA's occupational distribution and its wage implications by sex, without regard to race or ethnicity. The conclusions from our mul- tivariate analysis of the impact of race/eth- nicity on CETA experiences argue for this approach. In addition, analysis of two sexes and three racial/ethnic groups becomes too cumbersome for the resulting small gain in analytic detail. OCCUPATIONAL DESEGREGATION IN CETA Since FY 1974 millions of adult women have participated in CETA. In connection with the reauthorization of CETA in Octo- ber 1978, CETA regulations directed state and local CETA administrators to reduce sex stereotyping in employment and training. We only had data for October 1975 to Sep- tember 1978 for these analyses. Thus, we can only describe CETA's occupational seg- regation record prior to the introduction of the desegregation directive and cannot as- sess CETA's responses to this directive. At the same time, even prior to CETA's 1978 reauthorization, CETAespecially Title I was expected to improve the eco- nomic prospects of its clients. Since female- dominatec! occupations command lower wages than those of mixer! and maTe-domi- nated occupations, it is reasonable to look for evidence that CETA tried to train and employ women in mixed and male occu- pations. In describing CETA's occupational de- segregation record, we use the CETA's def- initions. In a male-dominated occupation fe- males constitute less than 25 percent of that occupation's labor force; in a mixers occu- pation, 25 to 74 percent; and in a female- dominated occupation, 75 percent or more. Table 16-4 shows the distribution of CETA jobholders among male, female, and mixed CETA jobs by sex and race. For FY 1976 to FY 1978, although only about 10 percent of the women in CETA jobs (work experience or PSE jobs) worked in male-dominated jobs, CETA placed about 25 percent in mixed occupations. Data published elsewhere show that CETA's occupational desegregation record for jobholders improved across the three fiscal years, the percentage of adult TABLE 16-4 Distribution of FY 1976-FY 1978 CETA Jobholders by Sex Composition of Occupation and Sex (percent) Sex and Race/Ethnicity Sex Composition of "Female Occupation . Male-dominated Female-dominated Mixed Total Male Hispanic Total Total White Black White 71.4 8.0 20.6 100.0 407,838 Black Hispanic N 10.8 64.1 25.1 100.0 100.0 10.9 62.8 26.3 401,176 256,073 11.9 64.4 23.7 100.0 115,261 6.2 74.4 19.4 100.0 29,842 71.1 8.3 20.7 100.0 605,484 70.3 8.2 21.4 100.0 150,568 47,078 70.1 11.0 18.9 100.0 SOURCE: Table 9, Berryman et al. (1981), p. 31.
Sex Composition of Occupation Male-dominated Female-dominated Mixed Total Total White 11.6 12.6 49.3 46.2 39.1 41.2 100.0 100.0 Black Hispanic Total JU, LINDA I. WAITE AND SUE E. BERRYMAN TABLE 16-5 Distribution of FY 1976-FY 1978 CETA Trainees by Sex Composition of Occupation ant} Sex (percent) Sex and Race/Ethnicity Female Male White 68.1 4.1 3.0 30.1 28.9 100.0 100.0 Black 63.3 6.8 30.0 100.0 Hispanic 57.1 6.5 36.4 100.0 9.5 55.8 34.7 100.0 9.1 55.8 35.1 100.0 65.9 N 56,264 38,030 13,359 13,792 104,828 74,169 16,867 4,875 SOURCE: Table 11, Berryman et al. (1981), p. 38. women employed in male-clominated CETA jobs increasing from 7 to almost 12, the per- centage in female-dominatec} CETA jobs de- creasing from 68 to 62, and the percentage in mixed jobs remaining stable (Berryman et al., 19811. Adult females showed slightly more distributional change across time than that for adult males, but neither sex shower] large changes. CETA's occupational desegregation re- cord in job training may be a better test of its desegregation success than is its record for jobholders. Since clients in job training presumably lack human capital in any spe- cific occupation, CETA's occupational as- signments should be less constrained by clients' prior occupational investments. More importantly, training can provide women with skills and credentials to enter male-domi- natec3 occupations. Table 16 5 shows the distribution of those in CETA job training among male-clomi- nated, female-dominated, and mixed occu- pations. Although CETA trained about the same percentage of women in male-domi- nated occupations as it employed in those occupations, it did train higher percentages in mixed occupations, reducing the per- centage in female-dominated occupations to a little under 50 percent. As the data in Table 16-1 suggest and data published elsewhere (Berryman et al., 1981) show, where training occurs (in a classroom or on the job) is clearly relater] to the sex composition of the occupation in which the person is trained. As noted earlier, the causal relationships between activity and occupa- tional assignments vary: an activity assign- ment may precede an occupational assign- ment, or vice versa, and in some cases both may be simultaneously determined by a third factor, such as title eligibility. Without ad- dressing causality, we can note that women in classroom training were 60 percent more likely to be trained in a sex-typical occu- pation and about 60 percent less likely to be trained in a mixed occupation than were women in on-thejob training. Although classroom training assignments reduced fe- male chances of being trained in a male- dominated occupation, the effects were not as great for this as for the other two occu- pational types. The data reveal that women in on-thejob training were more likely to be trained in mixed and male-dominated occupations pri- marily as a function of OlT's occupational mix. oTT contains much larger proportions of male-dominated and mixed occupations than does classroom training. Although women were substantially overrepresented in the female-dominated occupations in OIT, the smaller numbers of female-dominated occupational slots in OIT produced some occupational desegregation. These data in- clicate that if CETA increases women's OIT participation, it should simultaneously in- crease occupational desegregation for women.
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS 301 TABLE 16-6 Occupation of Last Pre-CETA Job by Occupation of CETA Job for Males and Females (FY 1976-FY 1978) (percent) Occupation in Pre-CETA Job Female Male- Female- Dominated Dominated 37.6 43.4 19.0 100.0 24 Occupation of CETA Job Male-dominated job Female-dominated job Mixed job Total N (000) Mixed Male Male- Dominated Fe male - Dominated 39.4 37.2 23.4 100.0 Mixed 54.0 7.6 38.4 100.0 6.8 75.8 17.4 100.0 9.9 44.8 45.3 100.0 138 83 84.0 4.2 11.9 100.0 279 27 114 - SOURCE: Tables 13 and 14, Berryman et al. (1981), pp. 36-37. Table 16-6 shows whether adult female and male CETA jobholders stayed in the same occupational type as their last pre- CETA job or moved to a new one. Thus, this table shows how much CETA changed participants' occupational patterns. About 75 percent of adult females in fe- male-dominated pre-CETA jobs entered fe- male-dominated CETA jobs. Of those who moved out of female-dominated pre-CETA jobs, more than two-thirds entered mixed CETA jobs. CETA retained less than 40 percent of adult females whose pre-CETA job was in a male-dominated occupation in their pre- CETA occupational type and placed more than 40 percent in female occupations. For females who had pre-CETA mixed jobs, CETA retained 45 percent in the same oc- cupational type and placed more than 40 percent in female-dominated occupations. Adult males had patterns similar but not identical to those of their female counter- parts; where CETA assignment altered oc- cupation it tendec! to move both males and females to occupations dominated by the same sex. A smaller percent of males than offemales shifted out of sex-typical pre-CETA jobs (16 and 24 percent, respectively). Males shifted out of sex-atypical pre-CETA jobs at almost the same rate as that offemales; they shifted out of mixed occupations at some- what higher rates. In sum, CETA changed the occupational type of proportionately more females than of males who had pre-CETA occupations typical for their sex. For those with pre- CETA mixed occupations or occupations atypical for their sex, CETA retained the same or a higher percentage offemales than of males in CETA occupations of the same type. However, CETA shifted only one- quarter of those females in female-domi- nated pre-CETA occupations into mixed or male-dominated occupations. It did not re- tain even half of those women in pre-CETA mixed or sex-atypical occupations in occu- pations of the same type and placed most of the changers in female-dominated occupa- tions, not mixed or male-dominated occu- pations. Finally, we can ask about CETA's record in meeting clients' occupational prefer- ences, as expressed in terms of its sex com- position.~3 The data on occupational pref- erences should be treated with caution. Participants answered the preference ques- tion after they had enrolled in CETA, and most had been assigned to an occupation. Their responses may be biased in the di- rection of their postenroliment occupational i3 The occupational preference data came from ques- tions on the CLMS that asked: "Did you want a certain kind of (~ob/job training) when you visited the man- power office?" [If Yes] "What was the (,ob/job training) that you wanted?"
302 LINDA J. WAITE AND SUE E. BERRYMAN TABLE 16-7 Distribution of Desired Occupation by Obtained Occupation for Male and Female CETA Jobholders (FY 1976-FY 1978) (percent) . . . ~ Desired Occupation Female Male Occupation of Male- Female- Male- Female- CETA Job Dominated Dominated Mixed Dominated Dominated Mixed , Male-dominated job 41.6 6.1 9.7 84.6 31.9 50.1 Female-dominated job 40.5 77.9 43.4 4.2 43.9 7.6 Mixed job 17.9 16.0 46.8 11.3 24.2 42.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 N (000) 22 140 81 253 26 104 SOURCE: Table 17, Berryman et al. (1981), p. 42. assignments. If they had no pre-enrolIment preferences, they may have responded to this question by naming their assigned CETA occupation. If they were assigned to an oc- cupation different from their preference, they may have accommodated to the discrepancy by modifying their original preference. Both of these potential biases would produce overestimates of the match between pre- ferred and actual assignment. As such, our data on the match between preferred and actual occupational assignments represent the maximum responsiveness of CETA to clients' preferences. In each fiscal year more than half of the adult female respondents indicated that they had had occupational preferences at the time of CETA entry. i4 For those women who ex- pressed preferences, a small but increasing proportion wanted male-dominated jobs across time (5 percent to 10 percent). An increasing proportion (from 26 percent to 35 percent) wanted mixed jobs, and a declining majority (from 69 percent in FY 1976 to 55 percent in FY 1978) wanted jobs in female- dominated occupations. Table 16-7 shows the CETA occupational distribution of adult females relative to their preferences at CETA entry. Fewer than half of the females who wanted male-dominated i4 Me percentages were 65, 57, and 59 for FY 1976, FY 1977, and FY 1978, respectively. or mixed jobs got them. Of the females who wanted and failed to get sex-atypical jobs, 69 percent ended up in femaTe-dominated jobs. Similarly, of the females who wanted but did not get mixed jobs, 82 percent ended up in female jobs. More than 75 percent of Me women who wanted female jobs got them; and of those who failed to get desired female jobs, almost three-quarters got mixed, not male, jobs. In sum, from FY 1976 to FY 1978 CETA employed or trained fewer than half of its female participants in male-dominated or mixed occupations. The percentages in- creased across fiscal years and were higher in on-thejob training than in CETA's cIass- room training or job services. Relative to their representation in the particular CETA service, females in on-thejob training were much more likely to be assigned to female- dominated occupations than were females in classroom training. OIT's better occu- pational desegregation record was attribut- able to the small number of female occu- pational slots in that activity. For women whose pre-CETA job had been a male or mixed occupation, CETA employed fewer than half in occupations of the same sex- composition type, shifting almost half of the "movers" into female occupations. For women whose pre-CETA job had been a female-dominated occupation, CETA shifted 25 percent to a mixed or male occupation primarily to the former. Finally, for women
who had occupational preferences at CETA entry, the match between preferred and ac- tual CETA occupation was much higher for those with preferences for female-clomi- nated jobs than for those with preferences for male or mixed jobs. WAGE IMPLICATIONS OF CETA OCCUPATIONS The low wages of female-dominated oc- cupations are one of the primary reasons for trying to desegregate occupations for women. From this perspective wages are the critical basis for judging women's occupational ex- periences in CETA. We examine their wages cluring CETA and the wages paid in the labor force as a whole for the CETA occu- pation in which they trained or were em- ployed. In-CETA Wages We assess sex differences in the CETA wage implications of female CETA occu- pational assignments in three ways: by 1- digit census occupational codes, the CETA service, and the sex-typicality of the occu- pation. Table 16-8 shows the real average hourly CETA wage by sex for the 1-digit census occupational codes. Without exception males earn higher hourly wages than do females in the same occupational category.~5 Table 16-9 shows the real average hourly wage by sex ant] CETA activity. Both sex and CETA activity affect CETA wages. If we look at wages by sex for the same CETA activity, males again get systematically higher wages than do females. The effects of CETA activity are the same for males and fe- males and, as data publishecl elsewhere show, for whites, blacks, and Hispanics i5 The large sample sizes make tests of signif~c~ce relatively uninformative. We examine wages for dif- ferences large enough to be significant substantively. 89. OCC1JPATIONAL DESEGREGATION IN CETA PROGRAMS 303 TABLE 16-8 Average Hourly Wage of CETA Occupations by Sex (FY 1976-FY 1979) (constant cloliars) N Occupation (000) N Males (000) Females Professional/ technical Managerial/ administrative Sales workers Clerical Crafts Operatives Transportation equipment operatives Laborers Service Average (136) 3.56 (39) (11) (98) (249) (121) 3.78 3.24 3.05 3.25 3.19 (63) 3.04 (380) 2.97 (326) 2.86 (1,422) 3.10 (133) 3.39 (21) 3.56 (12) 2.49 (537) 2.69 (20) 2.72 (43) 2.67 (7) 2.75 (32) 2.71 (217) 2.54 (1,023) 2.76 SOURCE: Table 37, Berryman and Waite (1982), p. (Berryman and Waite, 1982~. As the last col- umn of Table 16-9 shows, the rank order of wages by CETA activity is: public service employment ~ on-thejob training > work experience > classroom training. We noted earlier that females are more apt than males are to be funnelecl into Title I. Once in Title I, they are more apt than males are to be funnelecl into classroom training and work experience activities. Thus, a much larger percentage of CETA females than of CETA males are in the two activities (work expe- rience and classroom training) that receive the lowest CETA wages. Table 16-10 shows that women in CETA training and in CETA jobs received lower wages than men (lid in each of the three sex- composition occupational categories (Ber- ryman et al., 19811. The wage difference between the sexes was greatest for the fe- male-dominatec] occupations, less and about the same size in the male-dominated anal mixed occupations. Women in CETA job training received somewhat lower hourly wages if they trained in a female-dominated occupation than if they trained in either a male-cTominated or mixed
304 LINDA I. WAITE AND SUE E. BERRYMAN TABLE 16-9 Average Hourly Wage by CETA Activity and Sex (FY 1976-FY 1979) (constant dollars) N N N CETA Activity (000) Males (000) Females (000) Total Classroom training (210) 2.23 (297) 2.06 (507) 2.13 OJT (207) 3.22 (109) 2.70 (316) 3.04 Work experience (398) 2.51 (400) 2.38 (797) 2.45 Public service employment (794) 3.30 (476) 3.09 (1,270) 3.24 SOURCE: Table 44, Berryman and Waite (1982), p. 102. Occupation. However, training in a female- clominated occupation dic3 not reduce the wages of men relative to the wages of those training in male-clominatec! and mixed oc- cupations. Women in CETA jobs received the lowest wage rates in female-dominated occupations and the highest in mixed occupations. Al- though men in CETA jobs also received the highest wage rates in mixer! occupations, working in a female occupation clid not de- press their wages relative to the wages as- sociated with male occupations. However we categorize CETA occupa- tions by census code, CETA activity, or sex-typicality within each category wom- en's wages were on average about 90 per- cent of the men's wages. It is not clear how to judge this wage record. Although the av- erage difference between male ant! female wages in CETA was small, men consistently made higher wages than women did, and for two reasons the smallness of the differ- ence between them may be less impressive than it initially appears. First, CETA wages were subject to floors and ceilings, thus compressing the wage range for both sexes. Second, however we categorize CETA oc- cupations, participants in the same CETA activity or CETA occupation were probably more homogeneous even on unmeasured characteristics that affect wages than were members of an occupation in the general labor force. Post-CETA Wages We do not know the relationship between the occupation of the CETA job or job train- ing and that of participants' post-CETA jobs. However, if CETA clients train or work in occupations whose counterparts in the labor TABLE 16-10 Average Hourly Wage for CETA Trainees and CETA Jobholders by Sex Typicality of Occupation (FY 1976-FY 1978) (constant dollars) CETA Activity/Sex Male Female Typicality of Occupation Total White Black Hispanic Total White Black Hispanic - ~ . fannies Male-dominated 3.63 3.66 3.61 3.46 3.00 3.04 2.96 2.82 Female-dominated 3.52 3.64 3.39 3.46 2.79 2.77 2.78 2.93 Mixed 3.32 3.37 3.19 3.25 2.89 2.88 2.94 2.94 N (000) 103 70 16 13 56 37 13 5 Job Holders Male-dominated 3.34 3.43 3.11 3.25 3.12 3.21 2.95 3.18 Female-dominated 3.34 3.41 3.25 3.10 2.90 2.94 2.84 2.79 Mixed 3.53 3.60 3.36 3.42 3.35 3.39 3.18 3.45 N (000) 609 396 145 46 407 251 112 29 SOURCE: Tables 20 and 21, Berryman et al. (1981), pp. 50-51.
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS 305 TABLE 16-11 FY 1976-FY 1979 CETA Occupational Distribution by Sex and CETA Activity and the 1979 Unemployment Rates and Median Wages of Occupations in the Unsubsidized Sector Males Females Unsubsidized Sector 1979 Median Weekly EarningsC CETA CETA CETA CETA 1979 (Full-time Wage and Occupational Training Job Training Job Unemploy- Salary Workers) Categorya (percent) (percent) (percent) (percent) ment Ratesb (dollars) Professional/ technical 5.6 10.8 6.4 14.4 2.4 316 Managerial/ administrative 2.0 2.8 1.0 2.2 2.1 349 Sales workers 2.5 0.3 2.1 0.8 3.9 254 Clerical 6.1 7.2 53.3 53.5 4.6 195 Crafts 35.5 14.4 4.3 1.4 4.5 303 Operatives 28.7 4.0 11.9 1.4 8.4 211 Transportation equipment operatives 4.0 4.4 0.3 0.7 5.4 272 Laborers 7.7 30.1 1.0 3.5 10.8 206 Service 7.9 25.9 19.6 22.1 7.3 164 Total or average 100.0 100.0 100.0 100.0 5.8 244 N 375,246 1,189,725 332,945 872,307 a These are the 1-digit census occupational categories. They exclude three categories that do not occur in the CETA occupational structure: Farmers and Farm Managers, Farm Laborers and Supervisors, and Private House- hold Workers. b From Table A-23, Department of Labor (1980), p. 257. c From Table 704, Bureau of the Census (1980), p. 424. market as a whole have high unemployment rates, they should have less chance of cap- italizing on their CETA occupational expe- rience. If the labor market counterparts of their CETA occupations have low wages and CETA clients obtain a post-CETA job in the same occupation as their CETA occupation, their wages will be low. Table 16-11 shows how CETA males and females distributed across the 1-digit census occupational codes by CETA service (train- ing and jobs). It also shows the 1979 un- employment rates and median weekly earn- ings for these occupations in the labor market as a whole. The occupations with the highest 1979 unemployment rates were the opera- tive, laborer, and service occupations; those with the lowest median weekly wage rates were the clerical, operative, laborer, and service occupations. Females in CETA job training had about the same occupational distribution as that of females in CETA jobs. About 75 percent of the women in each of these activities fell into two occupations: clerical and service, both with low wages rates in the labor mar- ket as a whole. The service occupation also had relatively higher unemployment rates. Males in CETA job training had different occupational distributions than those of males in CETA jobs. Of those in CETA jobs, more than 50 percent fell into two occupations: laborer and service, both with low wage rates and relatively high unemployment rates. For males in CETA job training, almost two- thircls fell into two (different occupations: crafts and operatives. The former hall a moderate unemployment rate and relatively high wage rate; the latter, a relatively high unemploy- ment rate and low wage rate.
306 LINDA I. WAITE AND SUE E. BERRYMAN Thus, from FY 1976 to FY 1979 CETA employed most women in occupations with low wages in the labor market as a whole. CETA did not use training to alter the pro- portion of women in occupations offering relatively little economic security. CETA employed most men in two of the least eco- nomically secure occupations with relatively low median wages and high annual employ- ment rates. However, CETA used training to reduce the percentage of men in the four low-wage occupations from two-thirds to one- half. Conclusion Training or working in male-dominated or mixed occupations gave women higher CETA wages than those from training or working in female-dominated occupations. How- ever, CETA wages were consistently lower for women than for men in the same census occupation, in the same CETA service, or in the same sex-composition category. Of those in CETA jobs, CETA employed 80 percent of the women and 67 percent of the men in the four occupations whose un- subsidized counterparts had the lowest wages and/or high unemployment rates. For those in CETA training, CETA did not alter the percentage of women in lower wage occu- pations, but reduced the percentage of men in these occupations from 67 to 50 percent. However we judge CETA's occupational desegregation record, the bottom line of that record for women- their CETA wages ant] post-CETA economic prospects is not . . Impressive. SUMMARY AND POLICY IMPLICATIONS The analyses reported in this paper are useful as baseline information about women ant! CETA for the three years prior to CETA's 1978 reauthorization. We found that, relative to their eligibility for CETA, women were underrepresented in CETA's Titles I, II, ant! VI for all three fiscal years except in Title I for FY 1978. When we controlled on variables that should affect title and activity assignments, we found women overrepre- sented in training activities, especially in classroom training relative to OIT, and in income-transfer jobs relative to jobs de- signed to lead to permanent, unsubsidized employment. For all three years, female CETA partic- ipants were concentrated in female-domi- natec! occupations, although the concentra- tion was less among CETA job trainees than among CETA jobholders. CETA placed only about 40 percent of the women whose last pre-CETA occupation had been male-dom- inated or who had expressed a preference at CETA entry for a male-dominated oc- cupation. In both cases, for those not placed in a male-dominated occupation, about two- thirds were placed in sex-typical occupa- tions. However we categorized the CETA oc- cupation, within each category women's wages were about 90 percent of men's wages. Although the wage difference between men and women was not large, it was consistent, and for reasons discussed earlier in this pa- per, the smallness of the difference may be less impressive than it appears. In terms of their post-CETA prospects, about three- fourths of the women in CETA jobs and in CETA job training were employed or trained in occupations that paid low wages in the general labor market: service and clerical occupations. The policy implications of these data are not clear for three major reasons: 1. In connection with CETA's reauthor- ization, CETA prime sponsors were di- rected to reduce occupational sex segre- gation in CETA. Our analyses provide abase- line for assessing CETA's response to that directive but not its current occupational status. 2. One of the reasons for desegregating CETA occupations was to improve women's post-CETA wages. At this juncture we lack
OCCUPATIONAL DESEGREGATION IN CETA PROGRAMS _ . 307 analyses that show that being trained or em- ployed in male-dominated CETA occupa- tions positively affects women's post-CETA economic outcomes. If the effects are neg- ative for reasons substantially beyond the control of CETA, trade-offs between these two policy objectives reduced occupa- tional sex segregation and economic self-suf- ficiencyhave to be made. 3. The male-clominated occupations for which most female CETA participants will be eligible are blue-collar occupations. Structural changes in the American econ- omy- and the occupational consequences of these changes imply that we need to re-examine what occupations sex-typical or atypicalbest equip CETA participants for economic self-sufficiency. The avowed purpose of job programs, in- cluding CETA, is to improve the prospects of those who lack the skills to obtain ac- ceptable employment on their own. This means getting people jobs at decent wages. Moving women into jobs currently filled predominately by men is desirable to the extent that it serves this purpose. But a number of factors may decrease the utility of occupational desegregation as a means to the ends espoused by job programs. First, most women eligible to participate in job programs could enter white-colIar occupa- tions only through stereotypicaDy female jobs such as clerical work. The male-dominated jobs potentially available to them tend to be blue collar, primarily service, operative and, perhaps, crafts jobs. Many of these occu- pations show high rates of unemployment currently, and women seeking to enter them would face competition from large numbers of men. Second, the structural changes now taking place in the economy make unskilled and semiskilled blue-collar jobs especially susceptible to technical obsolescence. Third, little empirical evidence exists on the suc- cess of occupational integration as a mech- anism for improving the employment pros- pects of women. For these reasons, we argue that job programs for women should care- fi~ly assess their goals and the ways in which the sex composition of the occupation affects the chances of achieving those goals. REFERENCES Barrett, Nancy S. 1979 "Women in the Job Market: Occupations, Earnings, and Career Opportunities." P. 39 in Ralph E. Smith, ea., The Subtle Revolution. Washington, D.C.: The Urban Institute Press. Berryman, Sue E., and Linda J. Waite 1982 Hispanics and CETA: Issues of Access, Dis- tribution, and Equity. Unpublished working paper, Rand Corporation, Santa Monica, Calif. Berryman, Sue E., Winston K. Chow, and Robert M. Bell 1981 CETA: Is It Equitable for Women? N-1683- NCEP. Santa Monica, Calif.: Rand Corpora- tion. Bureau of the Census 1980 Statistical Abstract of the United States: 1980 (101st ea.) Washington, D.C.: U.S. Govern- ment Printing Office. 1981 Current Population Reports, Series P-20, No. 363. Population Profile of the United States, 1980. Washington, D.C.: U. S. Department of Commerce. Available from the U.S. Super- intendent of Documents, U.S. Government Printing Office. Chiswick, Barry, J. Fackler, June O'Neill, and Solomon Polacheck 1974 "The Effect of Occupation on Race and Sex Differences in Hourly Earnings." Pp. 219-228 in Proceedings of the American Statistical As- sociation. Washington, D.C.: American Sta- tistical Association. Department of Labor 1980 Employment and Training Report of the Pres- ident. Washington, D.C. Mirengoff, William, and Lester Rindler 1978 CETA: Manpower Programs Under Local Control. Washington, D.C.: National Acad- emy of Sciences. Sawhill, Isabel V. 1973 "The Economics of Discrimination Against Women: Some New Findings."Journal of Hu- man Resources 8: 383-396. "Discrimination and Poverty Among Women Who Head Families." Pp. 201-211 in Martha Blaxall and Barbara Reagan, eds., Women and the Workplace. Chicago: University of Chicago Press. Wolf, Wendy C., and Rachel Rosenfeld 1978 "Sex Structure of Occupations and Job Mobil- ity." Social Forces 563:823-844. 1976