1. Nicole M. Fortin Department of Economics and CIFAR University of British Columbia June 2008 European Society of Population Economists

2.  Over the last century, the most significant change in the U.S. labor market has been the remarkable growth in women's participation in the labor force.  From less than 5 percent at the turn of the 20th century, female labor force participation (FLP) among 18 to 65 year olds grew to over 70 percent in the mid 1990s peaking at 72 percent in 2000, before it began to retreat back to 70 percent in 2004 (and 2007).  Since the mid-1990s, there are also been a stabilization in female labor force participation (FLP) in many high FLP countries, such as Canada, Finland, Norway, Sweden, and the United Kingdom (see Fortin, 2005). 2

3. 3

4.  This stabilization or retreat in FLP has been characterized as “opting out” in the popular press (Belkin, 2003; Wallis, 2004; Story, 2005) and among sociologists (Cotter, Hermsen and Vanneman, 2007; Stone, 2007) who wonder whether we are witnessing “The End of the Gender Revolution” as an ideological movement.  Economists are more skeptic about the importance of the phenomena (Boushey, 2005; Goldin and Katz, 2007)  Women’s educational attainment has continued to rise, although their relative wages have not.  Their husband’s income has remained relatively unchanged, with an elasticity of income quite low (Blau and Kahn, 2005)  Demand-side (e.g. technological change, sectoral shifts) factors continue to favor women’s work. 4

5.  From the mid-1960s onwards, the spectacular rise in FLP has been shown to coincide with the “Pill” revolution (Goldin and Katz (2002) and Bailey (2006), which facilitated women’s access to higher education.  Yet, it happened in the same era as other changes, the Civil Rights Movement, the Sexual Revolution and more importantly the “Women’s Liberation Movement”, which proposed the new identity of “career woman”.  In recent years however, “feminism” has begun to carry negative connotations, intensive housewifery (à la Martha Stewart) and “intensive mothering” are on the rise.  The AIDS crisis, which peaked in the mid-1990s, tempered risky sexual behavior among the generation X, in particular.AIDS crisis, 5

6.  Using data from the 1972-2006 General Social Surveys (GSS) and from the National Longitudinal Survey of the Class of 1972 (NLS72) and an economic identity theory framework,  this paper contributes to the study of the impact of gender role attitudes on FLP in several ways, by 1) accounting for non-linear time-period, life-cycle and cohort effects, as well as a host of background variables, and 2) using a plausible instrumental variable strategy to dispel concerns about cognitive dissonance 3) corroborating 1) and 2) using longitudinal data to address the issue of reverse causality and possible misspecification of life-cycle and cohort effects. 6

7.  Using data from the 1972-2006 General Social Surveys (GSS) and from the National Longitudinal Survey of the Class of 1972 (NLS72) and an economic identity theory framework,  this paper contributes to the study of the impact of gender role attitudes  on FLP in several ways, by 1) accounting for non-linear time-period, life-cycle and cohort effects, as well as a host of background variables, and 2) using a plausible instrumental variable strategy to dispel concerns about cognitive dissonance 3) corroborating 1) and 2) using longitudinal data 7

8.  I find a strong negative impact of holding traditional gender role attitudes on FLP, robust to instrumentation and to a wide range of alternative hypotheses and a weaker positive impact of egalitarian attitudes.  Traditional gender role attitudes, whose decline bottomed out in the mid 1990s when the AIDS crisis peaked, are found to be the missing gender-specific factors that explain the differences in the evolution of male and female labor force participation, that remain after accounting for usual factors.  Importantly with regards to the “opting out” hypothesis, I find that these results are not driven by  the behavior of college-educated women  concerns about the welfare of children 8

9.  Impact of gender role attitudes on FLP  Levine (1993), Vella (1994), Fortin (2005), Fernandez and Fogli (2005)  Intergeneration transmission of gender role attitudes  Fernandez, Fogli and Olivetti (2004), Farré-Olalla and Vella (2007)  Dynamic macro-models with gender role attitudes  Fernandez (2007) and Fogli and Veldkamp(2007)  Cohort-effects – “Pill Revolution”  Goldin and Katz (2002), Goldin (2004) and Bailey (2006)  Identity and Gender  Akerlof and Kranton (2002), Goldin and Shim (2004), Goldin (2006) 9

10.  Akerlof and Kranton (2000, 2002, 2005) have proposed to incorporate one's sense of self as an important element of the utility function.  Benabou and Tirole (2006) have introduced competing identities that are competing for time or resources.  Here, these competing identities are v H the traditional identity of homemaker or housewife and v C the new identity of “career women”.  The model involves three stages: period 0, the choice of identity; period 1, the investment period; and period 2, reaping the rewards of identity, where identity is an asset- value pair {A,v}, yielding utility : vA. 10

11.  The woman starts with an initial endowment of each type of identity-asset, A H 0 and A C 0, which help her self-assess her type.  In period 0, she takes some initial identity-specific actions, a H 0 and a C 0 that will built up, with some uncertainty z,the identity-specific capital at rates r H 0 and r C 0, incurring costs c H 0 and c C 0.,  In period 1, the agent chooses her actions by reference to an imperfect recall of her initial type û, and by considering her actions from period 0 and by incorporating the anticipated utility, s 1 ûA 2, of the identity-asset to be realized in period 2, where s 1 is a saliency parameter. 11

12.  The individual estimating equation takes the form  Y it = α 0 + α 1 T + α 2 T 2 + α 3 A + γ 4 A 2 + Σ j δ j B j + β g G it + β x X it + ε it  where Y it is the outcome of interest  where T is time, A is age and B j are the 8 birth cohort categories  where G it are gender role attitudes that capture the saliency of traditional or egalitarian identities  where X iti individual characteristics that capture the identity-endowment (living in intact family, mother ever worked, religion at 16, etc.), identity-asset (education, children, married, divorced, etc.) variables  For simplicity, the model is estimated with a linear probability model, but corroborated by a Probit model 12

13.  Data  Descriptive Evidence  Time-period, life-cycle and cohorts effects in FLP  Measurement of gender role attitudes, other attitudes and divorce rates, number of children  Regression Results  Main results  Instrumental variables strategy  Alternative hypotheses (divorce, religious, social and political conservatism, ethnic and health factors) Alternative hypotheses  College-educated women vs. women with less than college College-educated women women with less than college  Falsification results for men Falsification results for men  Longitudinal analysis with the NLS-72 Longitudinal analysis with the NLS-72  Conclusion 13

14.  The main data are drawn the 1972 to 2006 General Social Surveys (GSS) conducted yearly (or bi-yearly) by National Opinion Research Center.  Each cross-section comprises 1372 to 2992 observations per year with a total of 23435 females and 19194 males between the ages of 18 and 65.  But the sample for which consistent gender role attitudes are measured comprises a subset of about 9000 women, because the same questions are not asked in each survey.  Data from National Longitudinal Survey of the High School Class of 1972 (NLS72) which follows the first post-Pill cohort (1954-55 birth cohorts) are also used. 14

15. 15

16. 16

17. 17

18. 18

19. 19

20. 20

21. 21

24.  In the GSS, out of a total of eight questions on gender role attitudes, only four are asked in the 2000s  Answers scaled 1 to n were rescaled 1 to 0 using the formula (n-k)/(n-1) where n is the number of categories and k is the categorical integer, but multiplied by 100 in Table 1. 24

25.  In the NLS72, there a total of ten questions: 5 on traditional views and 5 on egalitarian attitudes 25

26. 26

27. 27

28. 28

29. 29

31. 31

32. 32

33. 33

34. 34

35. 35

36. 36

37.  At -0.246 (0.022), the impact of traditional attitudes implies that the slight rise in average traditional attitudes from 0.373 in 1994 to 0.401 in 2006 would account for a [(0.401-0.373)*-0.246*100] 0.7 percentage point decline in FLP.  By comparison, the increase in years of schooling from 13.04 in 1994 to 13.12 in 2006 would have lead to an increase 0.18 percentage point.  Importantly, the introduction of traditional attitudes reduces the magnitude of quadratic term of the time trend from - 0.018 to -0.011, rendering it insignificant and comparable in point estimate to that of men. 37

38. 38

39.  When questions about gender role attitudes and labor market decisions are asked contemporaneously, this raises the issue of reverse causality or avoidance of cognitive dissonance (Akerlof and Dickens, 1982).  This issue is addressed using an instrumental variables strategy where the instruments are answers to questions about the respondents’ political views and attitudes towards sexual relations.questions  These variables are correlated with gender role attitudes and are thought to impact labor market decisions only through attitudes toward whether women should work outside the home or not. 39

40. 40

41. 41

42. 42

43. 43

44. 44

45.  I explored various alternative hypotheses 1) Increase in divorce rates and attitudes toward divorce 2) Social conservatism (attitudes toward premarital sex) 3) Political conservatism 4) Increased religiosity (church attendance, e.g. Glaeser and Sacerdote, 2007; bible inerrancy also tested, e.g. Sherkat, 2000) 5) Cultural background (42 dummies on ethnic ancestry, e.g. Fernandez and Fogli (2005), Zaiceva and Zimmerman (2007)) 6) Increasing rates of ill-health (morbid obesity, e.g. Cawley, 2004) 45

46. 46

47. 47

48. 48

49. 49

50. 50

51.  This paper provides compelling evidence that beliefs about gender roles are an essential element of the analysis of the evolution of FLP over the latter part of the twentieth century.  The empirical analysis of pooled cross-sectional data from the 1972-2006 GSS shows that traditional gender role attitudes are the missing gender specific factors that make gender differences in the evolution of labor force participation over the past 30 years fade away.  While the “opting-out" phenomena may be a terminology more appropriate for the popular press, concerns about the post-baby boom generations of women not be adopting the identity of ``career woman" with the same fervor as the pioneering baby-boom generations are not without foundation. 51

52.  While this paper solves one puzzle, it seemingly seems to open another one. How are gender role attitudes formed?  How can their evolution be explained?  To the extent that there is some intergenerational transmission of attitudes, is this bottoming out of traditional gender role attitudes a reflection of the relative demographic importance in the 1990s and 2000s of the Gen X generation, whose gender role attitudes may come from pre-World War II parents?  Will the attitudes of the following generations (post 1976 birth cohorts) be closer to that of their boomer parents? 52

53. 53

54. 54

55. 55 Note: Averages of strong agreement (1), agreement (2/3), disagreement (1/3), strong disagreement (0) of “It is much better for everyone involved if the man is the achiever outside the home and the woman takes care of the home and family”

56.  The exact questions are:  There's been a lot of discussion about the way morals and attitudes about sex are changing in this country. If a man and woman have sex relations before marriage, do you think it is always wrong, almost always wrong, wrong only sometimes, or not wrong at all? [VAR:PREMARSX]  What if they are in their early teens, say 14 to 16 years old? In that case, do you think sex relations before marriage are always wrong, almost always wrong, wrong only sometimes, or not wrong at all? ? [VAR:TEENSEX]  “We hear a lot of talk these days about liberals and conservatives. I'm going to show you a seven point scale on which the political views that people might hold are arranged from extremely liberal point 1 to extremely conservative point 7? Where would you place yourself on this scale?” [VAR:POLVIEWS] 56

57. 57

58. 58

59. 59

60. 60