Notes

Introduction

1. McGeehan 2004.

2. Ackman 2004b; Gibson 2004; Martinez 2004; McGeehan 2004.

3. New York Times, June 6, 2000; Wall Street Journal, July 29, 1999, July 30, 1999, September 24, 1999.

4. A. Fisher 1989; Herera 1997.

5. A. Fisher 1989.

6. Edelman (1990, 1992) has argued that organizations adopt equal employment and affirmative action policies to signal compliance to legal authorities, without much substance to those policies. In other words, the policies and procedures may be more symbolic than substantive.

7. Antilla 2002; New York Times, February 27, 1999; Wall Street Journal, June 22, 1999.

8. New York Times, February 27, 1999, C14.

9. New York Times, February 27, 1999; Antilla 2002.

10. New York Post, August 31, 1999.

11. Wall Street Journal, July 29, 1999, July 30, 1999, September 24, 1999; New York Times, August 3, 1999, June 6, 2000.

12. Ackman 2004b; Gibson 2004; Martinez 2004; McGeehan 2004.

13. McGeehan 2004.

14. Title VII protects both women and men from discrimination on the basis of sex. It does not protect against discrimination on the basis of sexual orientation or gender expression.

15. While sex may be a bona fide occupational qualification (BFOQ) for jobs that use specific reproductive capacities (e.g., gestational surrogate or sperm donor), this loophole has also been historically used in other occupations where its legitimacy is questionable (e.g., prison guard or flight attendant). In other words, gender stereotypes have been used to provide exemptions from EEOC sanctions under the BFOQ provision. Similar stereotypes do not provide legally permissible exemptions in cases of racial discrimination.

16. Griggs v. Duke Power Company 401 U.S. 424 (1971). The courts were largely favorable to employees during the 1970s, but reverted to favoring employers during the 1980s. In the Griggs decision, the court protected employees by interpreting disparate impact broadly. But the disparate impact doctrine began to crumble with decisions in Watson v. Fort Worth Bank & Trust (487 U.S. 977 [1988]) and Wards Cove Packing Co. v. Antonio (490 U.S. 642 [1989]). In Watson, the Supreme Court ruled that plaintiffs must demonstrate specific, isolated employment practices that caused the disparity, rather than placing the burden of proof on employers (as in the earlier Griggs decision). In Wards Cove, the Court ruled that employers need only produce sufficient evidence that their practices have some legitimate justification, and not that the practices producing disparate impacts were “essential” or “indispensable” to their business interests. In 1991 Congress overturned the precedent set by the Court in Wards Cove, with amendments to the Civil Rights Act. It codified the doctrine of disparate impact based on the earlier Griggs decision and placed the burden of proof firmly on the employer to demonstrate that the challenged practice is “job-related” and a “business necessity.” See Archer 2000; Hoff 1991.

17. Barron’s 1997.

18. Antilla 2002.

19. EEOC Guidelines on Discrimination because of Sex, 29 C.F.R., Section 1604.11 (a)(3).

20. Barr 1993; Fain and Anderton 1987; Glass 1988.

21. Goldstein 1988; Hoff 1991; MacKinnon 1989.

22. Eisenstein 1988; Finley 1993; Guthrie and Roth 1999a; Krieger and Cooney 1993; Magid 2001; W. Williams 1993. In 1978, Congress amended Title VII to prohibit discrimination because of pregnancy. Employers may not legally consider an employee’s pregnancy in making employment decisions, and they must treat pregnancy-related disabilities in a similar fashion to other disabilities that similarly affect an employee’s ability or inability to work.

23. Gruber 1992; Roth 1999.

24. Fletcher 1999; Foschi 1996; 2000; Reskin 2000; Ridgeway 1997; Ridgeway and Correll 2004.

25. Becker 1991; Hakim 2001; Mincer and Polachek 1980. For comprehensive reviews of human capital theories and other supply-side explanations for women’s lower earnings, see Blau and Ferber 1992; England 1992; Marini 1989.

26. Ryder 1965.

27. For more detail about the sample and methodology, see appendix A.

Chapter 1. The Playing Field: Wall Street in the 1990s

1. Ackman 2004a.

2. For example, see Becker 1971, 1991.

3. Wall Street Journal, May 9, 1999. The index dipped below 8,000 in the recession that began in 2000.

4. SIA Research Department 1998. NYSE broker-dealers account for about half of worldwide revenue and profits. The data presented are based on these firms because of data availability and because they reflects the core business of U.S. brokerage activity.

5. The ranking of investment banks was drawn from Eccles and Crane 1988.

6. SIA Research Department 1998.

7. Investment Dealers’ Digest 1998a.

8. Investment Dealers’ Digest 1998b.

9. SIA Research Department 1998.

10. Investment Dealers’ Digest 1997.

11. Wall Street Journal, October 21, 1998.

12. An informant in the industry estimated the recruitment rate at 15–20 percent. Confirming this estimate, the sampling frame for this research contained just less than 20 percent women (19.8 percent).

13. New York Times, February 24, 1998; Wall Street Journal, September 18, 1998, October 9, 1998, October 14, 1998, October 19, 1998, December 16, 1998.

14. SIA Research Department 1998.

15. Investment Dealers’ Digest 1998c.

16. Investment Dealers’ Digest 1998a.

17. I define “success” as working in corporate finance, research, sales, trading, or asset management and earning at or above the median in corporate finance for their graduating class (as defined by a survey of executive recruiters in Investment Dealers’ Digest 1998b), earning at or above the average estimated by Investment Dealers’ Digest (1998c) for equity research, or earning more than the sample median in sales and trading or asset management. I define those who worked in these areas but earned less than average as less successful. I also define those who worked in lower-paying areas like public finance or support functions, who left a higher-paying area for a lower-paying one, or who moved from a top Wall Street firm to a lower-tier financial firm or out of the industry as less successful.

18. I use the term “derailed” to describe workers who made choices that, deliberately or not, moved them off the fast track to success on Wall Street. Some derailments had more serious long-term effects than others. Some derailed professionals entered other positions where they might regain access to a path to success, although they would be behind schedule. But Wall Street contains a regular timetable for promotion and the pay that goes with it. From this perspective, derailment is an appropriate metaphor for the setbacks that some workers faced.

19. To protect the confidentiality of respondents, all names are pseudonyms. This includes not only respondent names but also names used in quotes and some place-names.

20. Investment Dealers’ Digest 1998b.

21. See Jacobs and Gerson 2004 for more detail on the frequency with which both men and women desire to effectively combine paid work and family life, and the desire of many workers to work fewer hours—especially those who work over fifty hours per week.

22. According to Investment Dealers’ Digest 1998b, the average in corporate finance for her graduating class in 1997 was $430,000, with most corporate finance workers earning between $300,000 and $600,000.

23. Note that these were relatively light hours for Wall Street but much higher than the average for the labor force as a whole. (See Jacobs and Gerson 2004.) The median among the Wall Street professionals interviewed in this study was sixty hours per week.

24. Because the men and women I interviewed, and Wall Street, were predominantly white, this book cannot offer a systematic analysis of race. The effects of race may be similar in some ways to the effects of gender, but how they are similar and how they are different cannot be explained using the experiences of seven nonwhite cases, especially given that these cases involved people of various races. I will mention race where it is relevant, but I am unable to systematically analyze it.

Chapter 2. Pay for Performance: Wall Street’s Bonus System

1. Eccles and Crane 1988.

2. A. Fisher 1989, 32; emphasis in original.

3. Chen and DiTomaso 1996; Smith et al. 2001.

4. These averages are the arithmetic mean rather than the median.

5. Blume, Siegal, and Rottenberg 1993; Kessler 2003; Lechner 1980; Matthews 1994. Until May 1, 1975, commissions for sales and trading were fixed at $0.26 per share. This dropped to $0.12 per share by the end of 1977, and the importance of commission revenue for securities firms continued to gradually decline.

6. A. Fisher 1989, 56.

7. It is worth noting that Wall Street workers viewed executive recruiters as useful sources of information on compensation amounts, especially given that Wall Street firms are very secretive about bonus pay.

8. U.S. Bureau of the Census 1998.

9. Investment Dealers’ Digest 1998b.

10. Blume, Siegal, and Rottenberg 1993; Hoisington 1976; Lechner 1980; Matthews 1994. Commission revenue constituted 61 percent of the revenues in major Wall Street securities firms in 1965, but was reduced to 50 percent in 1975, and to 17 percent by 1991 (Matthews 1994).

11. Blume, Siegal, and Rottenberg 1993; Jensen 1976; Kessler 2003; Matthews 1994.

12. Previous research has documented men’s greater tendency to ask for or demand resources and rewards, as well as the greater likelihood that they will receive what they ask for. See Babcock and Laschever 2003.

13. Kessler 2003, 122–23.

14. Investment Dealers’ Digest 1998c.

15. Some research analysts moved out of top Wall Street firms, usually to buy-side research positions. These workers were not included in the averages that I compare to Investment Dealers’ Digest (1998c) estimates. Including them, the range of earnings was $112,500 to $425,000, with a median of $193,750.

16. Since the market crash of 1987, markets are frozen after a certain volume of trades has been exceeded in a day in order to prevent trading frenzies caused by sudden fluctuations in the stock market.

17. In one case, after four years there were too many associates remaining in the group for all to become vice presidents. The firm’s response to this glut of employees was to lay off approximately half of the associates rather than hold them back from promotion. So these firms adhered quite diligently to the lockstep promotion schedule. There were a few cases where people were held back from promotion by one year, often because they had moved to a new area where they had limited experience.

18. In some firms, this title is director or principal.

19. Baron and Bielby 1980; Bird 1996; Blau and Ferber 1992; Dixon and Seron 1995; England and Farkas 1986; England 1992; Marini 1989; Jacobs 1989; Reskin and Roos 1990; Reskin 1993; Reskin, McBrier, and Kmec 1999.

Chapter 3. A Woman’s Worth: Gender Differences in Comparison

1. Only those who worked in the securities industry in 1997 were included, eliminating from the compensation data three men who had moved into other industries. Some of the men and women I interviewed left the industry after December 1997 but before the time of the interview. Because they were paid in the securities industry in 1997, I included them in compensation data and statistical analyses of compensation.

2. U.S. Census Bureau, 1998.

3. Becker 1971, 1991; Mincer and Polachek 1980.

4. See Hakim 2001.

5. Blau and Ferber 1992; Kilbourne et al. 1994; Marini 1989. Some have argued that this can occur as a consequence of “statistical discrimination,” whereby employers decide not to hire certain types of workers based on the average characteristics of the group to which they belong (Mincer and Polachek 1980). For example, if a majority of women prioritize family over paid employment and will interrupt their labor force participation for childbearing and child rearing, then employers may statistically discriminate against all female workers including the minority that is more work-committed than family-committed.

6. The model is an ordinary least squares (OLS) multiple regression model, which assumes that the background characteristics included in the model have a linear effect on total compensation. For more detail on quantitative measures and models, see appendix B. Undergraduate GPA and GMAT had no significant effects in any statistical models and are not presented in table B.2.

7. I asked each professional approximately how many hours he or she worked per week. Self-reported hours are expected to be inaccurate and somewhat inflated, although I expect that men and women inflated their weekly hours in similar ways. (See Jacobs and Gerson 2004 for evidence for this assumption.) In fact, one might think that there could be gender differences in the accuracy of self-reported hours, with women more likely to underreport hours because of the social stigma attached to excessive hours among women, especially those with children. But this potential difference in reporting would only widen the gender gap in earnings between men and women who work the same actual hours per week.

8. England 1992; Guthrie and Roth 1999a; Kilbourne et al. 1994; Petersen and Morgan 1995; Reskin 1993; Reskin and Roos 1990.

9. Baron and Bielby 1980; Bielby and Baron 1984; Dixon and Seron 1995; Guthrie and Roth 1999a; Halaby 1979; Reskin and Roos 1990.

10. Blau and Ferber 1992; England 1992; Epstein et al. 1999; Jacobs 1989; Kilbourne et al. 1994; Marini 1989. Using the metaphor of “revolving doors,” Jacobs argues that women experience a high degree of mobility over the life course among female-dominated, sex-neutral, and male-dominated occupations, and that this mobility ultimately reproduces patterns of sex segregation. So women may enter high-paying, male-dominated occupations, but many of them abandon these positions or move into part-time or lower-prestige work as they encounter glass ceilings and hostile environments. In highly male-dominated jobs, women may experience pressure as “tokens,” whereby they are highly visible and differences between them and the dominant group become exaggerated (Kanter 1977). Token women in male-dominated jobs may be highly constrained in their behavior and experience limitations to their further advancement or to increases in pay. Jacobs attributes the revolving door phenomenon to social control processes within workplaces, which can take obvious or subtle forms. Jacobs also suggests that discrimination by employers, harassment by supervisors and coworkers, and less access to informal organizational support contribute to a revolving door pattern for women.

11. Erickson, Albanese, and Drakulic 2000; Foschi 2000; Martin 2001; Pierce 1995; Reskin 2000; Ridgeway 1997.

12. Petersen and Morgan 1995; Treiman and Hartmann 1981. Treiman and Hartmann found that occupational segregation accounted for 35–40 percent of gender inequality when they controlled for 479 occupational categories. In a study of 16 occupations, Petersen and Morgan found that occupation-establishment segregation explained an average of 89 percent of within-occupation wage differences.

13. Budig and England 2001.

14. England 1992; Kilbourne et al. 1994.

15. Nelson and Bridges 1999. Also, see chapter 2 for criteria that influence bonuses on Wall Street and are not merit related.

16. England 1992; Reskin 1993; Reskin, McBrier, and Kmec 1999; Tomaskovic-Devey 1995.

17. This suggests that women may be more likely to be held back from promotion, although the reasons for that are unclear. The few cases where workers were held back from promotion usually involved a move into a new area where they had to learn new skills. But women were only slightly more likely to have this experience than men (one man and three women).

18. New York Times, February 27, 1999.

19. A. Fisher 1989; Herera 1997.

20. Rank affects earnings, although promotions to higher ranks on Wall Street are lockstep. Table 3.2 reveals that women were less likely to have attained the rank of vice president, while men were more likely to be above the vice presidential level. Because this would have influenced pay, I accounted for it in the statistical model in table B.3. Because the effects of having an undergraduate major in mathematics or engineering, previous experience on Wall Street, undergraduate GPA, and GMAT score were nonsignificant in all statistical models, they were not included in this regression analysis.

21. As table 3.1 illustrates, women on Wall Street were also slightly less likely than men to have spouses or children. At least at the time of the interview, many of these women did not have the kinds of family responsibilities that would make such trade-offs rational.

22. I put the average for all other characteristics into the regression equation and then computed the expected pay at 40, 50, 60, 70, 80, 90, and 100 hours for men and for women to form the two lines in the graph.

23. Nelson and Bridges 1999.

24. Acker 1990; Martin 2001.

25. Babcock and Laschever 2003.

Chapter 4. Making the Team: Managers, Peers, and Subordinates

1. An “opportunity context” is the environment that influences the opportunities that are available. This context enables or constrains individuals, leading to positive or negative economic outcomes. The opportunity context on Wall Street plays an important role in the career consequences of an individual’s credentials, family background, and work experience.

2. Acker 1990; Ferree, Lorber, and Hess 1999; Lorber 1994; Ridgeway 1997; Ridgeway and Correll 2004; Risman 1998.

3. Martin 2001; Reskin 2000, 2003.

4. In Kanter’s (1977) classic analysis, the proportion of a group within a work environment affects how its members are treated. When women are less than 15 percent of the group, they are isolated tokens and viewed as special for their group. At the same time, they are expected to represent their group. Sabrina fit this description as the only African American. But Kanter argues that as a minority increases to 20–30 percent, the majority feels more threatened and engages in more overt boundary-heightening activities. So when women reach 20–30 percent on Wall Street, men are likely to view them as a greater threat and engage in more boundary work to guard the industry as their territory. Women in Sabrina’s group represented 20 percent, which was similar to the proportion of women in Wall Street firms at the entry level.

5. Investment Dealers’ Digest 1998c. The average for research analysts without II rankings was $200,000–$350,000.

6. Twenty-eight men (88 percent) and 28 women (64 percent) said that their work groups were mostly male. Thirty men (94 percent) and 35 women (80 percent) said that their work groups were all white or mostly white.

7. Ridgeway 1997; Reskin 2000.

8. Martin 2001; McIntosh 1993. Subtle forms of discrimination might be best understood from the perspective of the dominant group, but dominant group members are often unaware of their privileges. Privileges usually operate in ways that are invisible to those they privilege, such that men think that sexism does not affect them because they are men, or whites believe that racism does not affect them because they are white (McIntosh 1993). On the other hand, women have a unique standpoint on how men’s interactions with each other in organizations unconsciously support and encourage men’s careers because those interactions have a visible impact on women’s experiences (Martin 2001).

9. Berscheid and Hatfield 1978; Fiske, Lin, and Neuberg 1999; Foschi, Lai, and Sigerson 1994; McPherson, Smith-Lovin, and Cook 2001; Reskin 2003; Ridgeway 1997; Ridgeway and Walker 1995; Smith-Lovin and McPherson 1993; Webster and Foschi 1988.

10. Erickson, Albanese, and Drakulic 2000; C. Fisher 1982; Ibarra 1997; Kanter 1977; Marsden 1987; Tsui and O’Reilly 1989.

11. Sexuality, especially the assumption of heterosexuality, is also infused in this phenomenon. Acker (1990) discusses how organizations not only are gendered but also contain assumptions about sexuality. The assumption of heterosexuality in Wall Street firms implies that mixed-sex interactions are always open to being interpreted as sexual while same-sex friendships tend not to be subject to this interpretation. The greater level of comfort with same-sex peers and subordinates is then also connected to assumptions about heterosexuality on Wall Street.

12. See table 2.1.

13. In comparison, 66 percent of men indicated that they had a mentor. Ninety percent of their mentors were men.

14. Martin 2001.

15. Budig 2002; Fairhurst and Snavely 1983; C. Williams 1992, 1995; Yoder 1991, 1994.

16. Berger et al. 1977; Ridgeway and Walker 1995; Webster and Foschi 1988. Gender effects are a specific case of the general tendency to give the benefit of the doubt to higher-status individuals while holding lower-status individuals to a higher standard. In this specific case, gender is a status characteristic, and men are accorded higher status than women.

17. Foschi, Lai, and Sigerson 1994; Foschi 1996, 2000.

18. Analysts in investment banking are the lowest-ranking employees. They are usually young college graduates who work for two years to gain some experience before going to graduate business school, and should not be confused with research analysts. In research, “analyst” is the general title of professional employees, who usually completed an MBA. In investment banking, analysts perform the most menial tasks for senior bankers who bring in client business.

19. Babcock and Laschever 2003; Bybee, Glick, and Zigler 1990; Clancy and Dollinger 1993; Cross and Madson 1997; McCrae and Costa 1988; McGuire and McGuire 1982.

Chapter 5. Bringing Clients Back In: The Impact of Client Relationships

1. Internal clients are investment bankers, salespeople, and traders. Research analysts often advise these other areas, providing recommendations of prospective investment banking clients in their industries and recommending buy, sell, or hold on companies’ stocks.

2. Consumer preferences are important for any industry that delivers a service. A substantial body of research in economics has examined consumer-driven discrimination (Andersen and La Croix 1991; Arrow 1972; Becker 1971; Caplow 1954; Crofton 2003; Holzer and Ihlanfeldt 1998; Kahn 1992; Kahn and Sherer 1988; Nardinelli and Simon 1990; Neumark, Bank, and Van Nort 1996). Like research on employer discrimination, this work usually emphasizes deliberate discrimination rather than subtle preferences that privilege some workers.

3. Strong relationships were also advantageous from the client’s perspective, since a stronger relationship offers a greater ability to trust the service provider and reduces the risks of opportunism in the course of a transaction (Baker 1990; Portes and Sensenbrenner 1993). Compared to shopping for the best price among competing financial institutions, trust provides advantages that offset any additional costs of services from a firm with an existing relationship.

4. I was unable to interview clients, so I rely on finance professionals’ interpretations of their client interactions. Since relationships are interactive, each party’s interpretations affect the character of the relationship itself. Given that information from the perspective of clients was not available, I must use workers’ interpretations to analyze the importance of client relationships, which is an important piece of the story of gender inequality on Wall Street.

5. Erickson, Albanese, and Drakulic 2000; Tsui and O’Reilly 1989.

6. The exception to this in Maureen’s remark was children. But discussing children could have very different career implications for women than for men. Children may be a form of positive social capital for men, defining them as serious and stable workers. For women, on the other hand, talking about their children could give clients the impression that they were not focused on their careers and were unprofessional. I discuss the effects of children for men and women in detail in the next chapter.

7. Bernstein 2001; Erickson and Tewksbury 2000.

8. Acker 1990; Martin 2001; Threadgold and Cranny-Francis 1990. Some might argue that clients could get an extra sexual thrill from attending strip clubs with female bankers, but this would be counterproductive to their image as knowledgeable professional experts. Clients might sexualize women who wanted to go to strip clubs, but they probably would not view them as more competent or trustworthy.

9. Average pay in public finance was $260,000. For a comparison with other areas, see figure 2.1.

10. Reskin and Roos 1990.

11. I would argue that employers hire men first for traditionally “male jobs,” and women first for traditionally “female jobs,” rather than always expressing a preference for men. But all jobs on Wall Street are historically male jobs; thus, men are at the top of the industry’s labor queue.

12. This informant was a personal contact, not a respondent.

13. Trentham and Larwood 1998.

14. This assumption is historically false, as there has been substantial conflict between Asian cultures that continues today.

Chapter 6. Having It All? Workplace Culture and Work-Family Culture

1. E.g., Budig and England 2001; Waldfogel 1997.

2. To construct this figure, I ran multiple regression models that calculated the effects of background characteristics (marital status, undergraduate major, and previous experience on Wall Street), average hours per week, rank (below or above vice president), and firm prestige on total compensation in 1997 for each parental status. I then computed the estimated pay by entering the group mean for each characteristic into the regression equation. The numbers presented in figure 6.1 reflect group differences when the effects of these variables are accounted for. I did not include area of the firm largely for empirical reasons. The number within each parental status group was too low to account for all of these influences simultaneously. Also, mothers often changed areas because of work-family conflict or discrimination on the basis of pregnancy or motherhood, so that area appeared to often be a consequence of the processes that I explore in this chapter rather than a cause. Raw averages for each group exhibit a similar pattern, as shown in table 6.2. In an analysis of variance, the difference in raw means across groups was statistically significant.

3. Hochschild 1997; Schor 1991. Hochschild’s analysis emphasizes workers’ incentives to spend more time at work and disincentives to spend time at home. Schor is more ambivalent about workers’ preferences to work, viewing managers as the more important force in the trend toward long hours. She argues that employers pay a premium to overwork employees, which most workers accept even if many might prefer to work less.

4. Clarkberg and Moen 2001; Harrison and Bluestone 1988; Jacobs and Gerson 2004. These scholars argue that jobs have increasingly been divided between high-paying jobs that require long hours and low-paying jobs that do not provide full-time work. As a result, workers at the top of the labor market, like Wall Street professionals, are overworked while many at the bottom are underemployed.

5. Bailyn 1993; Blair-Loy 2003; Blair-Loy and Wharton 2004; Jacobs and Gerson 2004.

6. Bernard 1981; Gerson 1985; Skolnick 1991; J. Williams 2000. This gendered division of labor fails to fit the reality of most contemporary families in the United States, in which women are in the labor force. Single-parent homes make up approximately one-third of families with children under eighteen.

7. DiMaggio and Powell 1991; Friedland and Alford 1991; Jepperson 1991.

8. Blair-Loy 2003; Clarkberg and Moen 2001; Moen 1992.

9. Clarkberg and Moen 2001. Preferences for one or both partners to work part-time were very common in their study of couples.

10. Clarkberg and Moen 2001, 1133.

11. Blair-Loy and Wharton 2004. Blair-Loy (2003) defines the “devotion to work schema” as an ideal that is shared by workplace organizations and many workers, in which work requires an intense commitment that precludes much involvement in other pursuits such as caregiving.

12. Blair-Loy 2003; Hays 1999. Blair-Loy called this the “family-devotion schema.”

13. This reflects another cultural ideology that is dominant in the United States: individualism. According to this ideology, individuals have control over their destiny through the choices they make, and many social structural forces that affect people’s outcomes are denied, ignored, and rendered invisible.

14. Nick’s experience fits well with Hochschild’s (1997) argument that workers prefer to spend time at work than at home. Hochschild claims that this preference is common because contemporary workplaces “value the internal customer” and provide rewards, encouragement, and a warm social environment. At the same time, she claims that as family life has become increasingly complex and homemakers are less common, the home requires work and involves emotional stress that many people prefer to avoid.

15. Nine of the seventeen fathers were extremely devoted to their careers and made large sacrifices in their personal lives. They all had homemaker wives, as did five other men who worked fewer hours and made time with their families a higher priority.

16. Wall Street workers were unusual in two noteworthy ways. First, only one of the workers I interviewed was a single parent–one mother was divorced. Second, the breadwinner-homemaker family has declined to approximately 10 percent of families within American society, largely due to economic necessity. But because Wall Street workers earned enough to support their families without a second income, the breadwinner-homemaker family was common in this industry. Men on Wall Street all married women who earned less than they did, and most married women who worked in low-paying occupations (e.g., teacher, flight attendant, social worker). Three men married another professional, but their wives ended their careers to become homemakers.

17. U.S. Bureau of the Census 2002. Among women aged 15–44, 44 percent of are childless. Among American women aged 25–34 with postgraduate degrees, the group most likely to be childless or delay childbearing, 42 percent were childless.

18. Gerson 1985; Hays 1999; Hertz 1997; Risman 1998; J. Williams 2000.

19. Same-sex relationships and same-sex mating markets might destabilize this process, but only one woman had a same-sex partner and this couple did not have children. Her partner had a similar job in the securities industry, so her situation resembled that of many of the women in relationships with professional men. Without more cases, I cannot determine if same-sex couples on Wall Street would have different dynamics vis-à-vis parenting roles.

20. This represents eleven of the twenty-six women with partners. According to Wall Street standards (see chapter 1, note 17), twelve women (27 percent) were highly successful. Among them, four were married to men who earned even more than they did.

21. For simplicity I use the word “men” here because all of the mothers were heterosexual.

22. Hays 1999; Hertz 1997.

23. This figure and table 6.2 use raw averages for pay and hours for each gender and parental status, rather than the estimated pay figures presented in figure 6.1. The estimated pay in figure 6.1 already accounts for differences in hours.

24. Childless men were much more likely to arrange a group dinner in the office or to horse around while they worked. This time was likely included in their estimations of their hours because it constituted face-time even though it was not productive time.

25. Recall from chapter 3 that hours were not significantly related to pay after accounting for all background characteristics, area, rank, and firm.

26. Gutek 1985.

27. Regarding litigators, see Pierce 1995.

28. Acker 1990; Martin 2001.

29. Budig and England 2001. After assessing a number of other theories for the wage gap, including explanations based on education, training and work experience (or “human capital”), and sex segregation, Budig and England argued that the residual wage difference between mothers and nonmothers could be a result of productivity differences, employer discrimination, or some combination of both.

30. In the remaining case, Kim left the industry after she got married but before she had children. Her primary reason for leaving was to relocate for her husband’s job. She also expected have children soon and to be a full-time mother for at least a few years.

31. Becker 1971; England 1992; Marini 1989. “Supply side” generally refers to characteristics that workers bring to the labor force and the choices that workers make. “Demand side” refers to employers’ preferences, tastes, or requirements, which affect the jobs available to workers and the terms of those jobs.

32. Martin 2001.

33. In fact, it could be argued that workers with a better work-life balance are likely to be healthier and happier, and therefore more effective when they are at work.

Chapter 7. Window Dressing: Workplace Policies and Wall Street Culture

1. Blair-Loy and Wharton 2004; E. Kelly 1999. Kelly refers to this as the “business case” for work-family policies.

2. Guthrie and Roth 1999b.

3. See Dobbin et al. 1993; Edelman 1990, 1992; Guthrie and Roth 1999b; Sutton et al. 1994; Sutton and Dobbin 1996.

4. Blair-Loy and Wharton 2004.

5. Fifty-three percent of workers (40) said that their firms were equipped to manage sexual harassment effectively, including 56 percent of men (18) and 50 percent of women (22).

6. The belief that reporting sexual harassment can produce retaliation and greater difficulties in the workplace is common and there is evidence to support it. See Fain and Anderton 1987; Fiske and Glick 1995; Livingston 1982; Roth 1999; Tangri, Burt, and Johnson 1982.

7. Lewis 1989; Wolfe 1987.

Chapter 8. Beating the Odds: The Most Successful Women

1. As success is defined in chapter 1, note 17.

2. The median was between $580,000 and $625,000, or $602,500.

3. Investment Dealers’ Digest 1998b.

4. Kanter 1977.

5. Chen and DiTomaso 1996; Kanter 1977; Reskin 2003.

6. Kanter 1977.

Chapter 9. The Myth of Meritocracy: Gender and Performance-Based Pay

1. This person was an informant who contacted me because she had heard about my research, not someone I interviewed.

2. See Becker 1971 for the most influential statement of this point.

3. Nelson and Bridges 1999. See also Frank 2004 on the elevation of free market ideology in American cultural rhetoric.

4. Kanter 1977.

5. Collins 2002; England 1992; Nelson and Bridges 1999; Phillips and Taylor 1980; Steinberg 1990.

6. Fletcher 1999.

7. Gutek 1985.

8. See Nelson and Bridges 1999 for further arguments on the fetishizing of market processes in American organizations and law, and Frank 2004 on the same process in American culture and politics.

9. England 1992; Reskin, McBrier, and Kmec 1999; Tomaskovic-Devey 1995. There is substantial evidence that jobs become devalued when they feminize. The causal direction for this relationship is debatable. Jobs may feminize because they are becoming less attractive to men, or the influx of women itself may lead to a decline in pay and prestige.

10. Blair-Loy and Wharton 2004; Kelly 1999.

Appendix A. Methodology

1. Eccles and Crane 1988. Eccles and Crane wrote a comprehensive study of investment banking organizations in the 1980s. I used their rankings to determine the pool of top firms that I would include in this study.

2. E. J. Kelly 1985; Walter 1985.

3. I attempted to contact anyone with a gender-ambiguous first name in order to ascertain which sampling frame to put them in.

4. They were presented with a list of specific names of classmates who had been selected. This proved to be a useful strategy for locating respondents.

5. Three respondents refused to permit tape recording. In those cases, I took detailed notes during the interview and transcribed it as soon as possible after the interview’s completion.

Appendix B. Quantitative Measures and Models

1. Bollen and Stine 1990. I also used correlation coefficients and collinearity diagnostics to test for multicollinearity. The variance inflation factors and condition index for the OLS models suggested that multicollinearity was not a problem. Regression diagnostics also revealed no outliers that would have skewed the results of the models.

2. Halvorsen and Palmquist 1980.

3. The same formula was used to calculate the change in the coefficient from the model in table B.2 to the full model in table B.3.

4. Cohen 1988.