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.
Commentary PAMELA STONE CAIN Chapters 2, 3, and 4 view segregation from three different vantage points. Beller pre- sents an analysis of aggregate occupational trends. Bielby and Baron explore the or- ganization of jobs within firms. Rosenfelc! looks at the occupational shifts of job chang- ers. The resulting picture is a confusing one. Beller emphasizes the decline in segrega- tion from 1960 to 1981, Rosenfeld the ex- istence of significant (i.e., nontrivial) move- ment across the boundaries of sex-typed occupations. In contrast, Bielby and Baron highlight the finding that segregation is vir- tually complete across jobs and firms of widely varying characteristics. From their results, it appears that only about 10 percent of workers are employed in integrated job ti- tles, and even in those jobs they often work at different sites or with different clients. Moreover, Bielby and Baron found no de- cline in job segregation over a period roughly comparable to the one studied by Beller. How to reconcile these results? In part, the contradictions can be attributed to inter- pretation. Although the segregation index declined at an ever-increasing rate, Beller finds that the resulting level of segregation was still very high: in 1981 approximately 87 60 percent of workers of either sex would have had to change occupations in order to achieve identical male and female distri- butions. Indeed, over a 20-year period, the index dropped by only 5 percentage points. Rosenfeld finds that over a 1- to 5-year period, 10 to 15 percent of the individuals in her sample of job changers moved from an occupation typical of their sex to one that was atypical; 60 to 70 percent moved from an atypical to a sex-typical occupation. Thus, over a relatively short period of time, only about 15 to 30 percent of job changers failed to cross the sex-typed boundary. These fig- ures indicate a movement of individuals across segregated occupations that is not imme- diately obvious given the level and con- stancy of segregation indices for the aggre- gate distribution. These results, however, like those of Beller, are less cause for op- timism when one realizes that fewer than 5 percent ofthe 1973 Current Population Sur- vey sample changed jobs. Their behavior, then, represents a flux at the margin. Most workers were stayers not movers, and, judg- ing from the other papers in Part I, they stayed in sex-typical jobs. The inconsistency in results between Beller
88 PAMELA STONE CAIN and Rosenfeld, on one hand, and Bielby and Baron, on the other, is also attributable to different units of analysis. Beller's and Ro- senfeld's use of relatively aggregated occu- pations underestimates the degree of seg- regation, given the considerable heter- ogeneity of jobs within even fairly detailed three-digit census categories. Beller and Ro- senfeld recognize this but nonetheless adopt a more optimistic, "half full" interpretation of their results. In light of Bielby and Bar- on's findings and the considerations cited above, a "half empty" interpretation might be more in order. This is not to deny that Belier and Rosen- fekI found evidence of change, however in- cremental and sIow-moving. Moreover, each found that sex-typed patterns were most yielding among younger workers. Both also found that the clecline in segregation was con- centrated among certain occupations, espe- cially those in the professional, managerial, service, and health care sectors. Other sec- tors, particularly skilled crafts, appear to pre- sent formidable barriers to women's entry. Although Bielby and Baron found little evidence of change, a case-by-case analysis of those companies that did move toward! greater integration led them to tentatively identify several factors responsible for, or at least facilitative of, such change. High on their list are rapid growth in company size and technological changes. To a lesser ex- tent, the presence of a sizable female work force was also helpful. Surprisingly, federal intervention in the form of contract com- pliance regulations during the early period appears to have had no impact on segrega- tion. By enabling an empirical assessment of the correlates (if not the causes) of seg- regation, Bielby and Baron's unique firm- level focus contributes immensely to a bet- ter understanding of the overall downward trend in segregation that Beller documents. Beller attributes the decline primarily to equal employment opportunity legislation and enforcement activities, citing the en- actment of Title VII and the Equal Rights Act early in the period under study. Putting aside questions about using time-ordering as a basis for attributing cause, the credi- bility of this interpretation is undermined by what we know about the slowness and inefficiency of the enforcement process and by the limited scope of some federal en- forcement efforts, especially contract com- pliance. Bielby and Baron's cross-sectional and longitudinal analyses suggest that more im- portant than regulatory efforts are larger structural changes in the economy that lead to the creation of new types of work, to new forms of work organization, and/or to changes in employee-employer relationships. They find less pronounced segregation in estab- lishments engaged in nonmanual service work, with unspecialized job structures and an absence offormal bargaining agreements. Larger firms with well-structured internal labor markets as well as very small firms were extremely segregated. Projecting these results and assuming that the factors Bielby and Baron identify are causes not correlates of segregation, there is reason to believe that segregation could decline as the national economy shifts from a manufacturing to a service base and the number of workers covered by collective bargaining drops. Countering these trends, however, are trends toward larger firm size and greater bureaucratization. Unfortu- nately, the main effects of each factor can be disentangled only in multivariate analy- ses and then only if the individual factors are not highly correlated with one another. Ultimately, it is the complex interplay of these factors that will determine the direc- tion of segregation in the 1980s, and it is difficult to predict this outcome given our current level of understanding. Certainly, a continuation of the downward trend that Beller documents cannot be taken for granted, given the scenario that Bielby and Baron depict. Confounding these structural changes that may affect segregation are the changes noted
COMMENTARY 89 by Rosenfeld in women's family and house- hold responsibilities, which have been brought on largely by inflation, unemployment, and divorce. As their paychecks become increas- ingly critical to household survival, women can even less afford, if indeed they ever could, cr , ,, women s wor <. Given that naturally occurring trends have opposing implications for segregation and that many previous interventions appear to have had little effect, what policy implica- tions can be drawn from these papers? Bielby and Baron's contention is persuasive that, with such widespread segregation, any in- tervention is warranted. What, then, should be the focus of intervention? Beller views the sex composition of occupations as es- pecially amenable to manipulation, presum- ably through the use of quotas and various affirmative action strategies. Enhancing the success ofthis strategy would be the changes in younger workers' job preferences that are already taking place. Bielby and Baron's results suggest that the solution may not be straightforward. Fundamental forms of workplace organiza- tion appear to be in need of alteration if segregation is to abate. Some of the crucial factors they identify (e.g., firm size, degree of job differentiation, and unionization) are outside the scope of government interven- tion as typically conceived. If their analysis is correct, the greatest hope for future in- tegration may lie not in the public policy domain but in the hands of workers them- seIves, through either more enlightened collective bargaining or other mechanisms of workplace democracy. The complex or- ganization of the contemporary workplace appears to play a major role in maintaining segregation. This organization can perhaps be changed only from within. Women moving into professional and managerial positions may be a catalyst in transforming the traditional organizational structures that have functioned so long to exclude them. As other papers in this vol- ume suggest, however, change will not be accomplished easily or without resistance (see, for example, Boos and Reskin). Turning to the research implications of these papers, Bielby and Baron's paper re- minds us of the distinction between re- search questions and researchable ques- tions. On the face of it, such guiding questions as "Why is there job segregation?" and "What causes it?" are plausible until it is recog- nized that the level of job segregation may not vary substantially. As a virtual constant, job segregation defies social science, as all our empirical methods determine cause and effect through covariation of variables. Thus, in the absence of variation, we cannot really dd . . .. . explam segregation. This predicament may direct us to explore phenomena that do vary to see what they can tell us about sex segregation. For ex- ample, particular occupations and jobs have changed sex type, i.e., are no longer mainly female or male. Newly emerging occupa- tions such as computer programming, as Beller finds, are becoming rapidly identified with one sex. Bielby and Baron's work also points out that, although the level of seg- regation does not vary much from firm to firm, specific jobs may be female in one firm and male in another. These examples lead to interest in historical or comparative in- quiries into the determinants of the sex com- position of a job or job family. A new research question dictates changes in methods. Ibe importance affirms in Bielby and Baron's work establishes the importance of using jobs rather than occupations as the unit of analysis and the concomitant need for firm-specific data. This is not merely be- cause the use of job data uncovers more segregation, but because job-level analysis picks up organizational context and varia- tions in hiring and allocation that occupa- tional-level analysis does not. Firms appear to exert a strong effect on occupations, and disregarding their context may obscure more than it illuminates. For example, it is dif- ficult to know what the occupational shifts Rosenfeld measured represent, especially as
go PAMELA STONE CAIN we do not know whether they were volun- tary or involuntary moves. Moreover, the high rate of movement across sex-typed oc- cupations may be misleading. A move to a sex-atypical occupation is not necessarily a move to an integrated job. Focusing on jobs, not occupations, raises problems of data availability. Few individ- ual-leve! surveys have any information on respondents' jobs or firms. At firms them- seives, it is often difficult to obtain coop- eration and access. These problems point to the need for new data collection efforts to supplement existing surveys. More broadly, the difficulties Beller encountered in com- piling a series of data on occupational seg- regation should alert us to the need for im- proved national statistical reporting systems that would enable us to monitor this critical indicator of sex equity. But prospects for better data at the national level are not en- couraging, given the current political cTi- mate. The agency within the U. S. Depart- ment of Labor, for example, that collected the data Bielby and Baron used in their anal- ysis has now been all but shut down. Bielby and Baron's paper also under- scores the need for new analytic strategies. Their survey analysis complements case studies by Rosabeth Moss Kanter (Men and Women of the Corporation, Basic Books, New York, 1977) and others, and, in their careful attention to selected firms (outliers), they blend case study and survey method- ologies. The richness oftheir analysis should settle the debate between those who ad- vocate one method over another. Clearly we need both, especially in this area, because of the problems of obtaining valid survey data on jobs and firms from either workers or employers. Finally, how do these studies enlighten the major debate of whether segregation is the result of employee choice or employer discrimination? These papers offer evidence for both positions. Within firms, Bielby and Baron document the existence of powerful mechanisms of control that would support an employer-side explanation. Among work- ers, Rosenfeld shows considerable circula- tion across sex-typed occupations. More- over, she finds some evidence that male employees are making rational choices in their avoidance of lower-paying, female- dominated occupations as their family re- sponsibilities increase. The question of cause employee or employeris un- doubtedly too simply framed. The different levels of analysis and seemingly contradic- tory findings of these papers highlight the complexity of the etiology of job segrega- tion.