Portal de Periódicos da CAPES

PlusÇa Change?1 evidence on global trends in gender norms and stereotypes

2007; Taylor & Francis; Volume: 13; Issue: 2 Linguagem: Inglês

10.1080/13545700601184880

1466-4372

Stéphanie Seguino,

Intergenerational and Educational Inequality Studies

Abstract Abstract Gender norms and stereotypes that perpetuate inequality are deeply embedded in social and individual consciousness and, as a result, are resistant to change. Gender stratification theories propose that women's control over material resources can increase bargaining power to leverage change in key institutions, prompting a shift to more equitable norms. By extension, policies that promote women's paid employment should serve as a fulcrum for gender equitable change. Is there any evidence to support this hypothesis? Investigating this requires a means to capture gender norms and stereotypes. The World Values Survey provides just such a mechanism because it contains a series of gender questions that span a twenty-year period and includes respondents from more than seventy countries. This paper uses that survey's data to analyze determinants of trends in norms and stereotypes over time and across countries, and finds evidence that increases in women's paid employment promotes gender equitable norms and stereotypes. Keywords: Economic growthemploymentgender ideologygender norms and stereotypesgender rolesglobalization JEL Codes: A14, J16, J21 Acknowledgments I am grateful to three anonymous referees for their incisive comments. I would also like to acknowledge the impressive research assistance of James Lovinsky and Tarik Yeasir. Irene van Staveren, Heather Antecol, Johannes Jütting, and Nancy Folbre generously provided useful comments as well. To all, thank you. Notes 1“Plus ça change, plus c'est la même chose” is a French idiomatic expression that literally translates as “The more things change, the more things stay the same.” Metaphorically, it can be translated as “History repeats itself.” Or, to expand that, “That which has been is what will be, and that which has been done is that which will be done. So, there is nothing new under the sun.” 2Individuals may have multiple goals and diverse identities that sometimes clash. Gender identity, however, has been cited as being of singular importance in shaping individual actions and societal pressures. For analyses of how identity affects economic behavior, see Nancy Folbre (1994 Folbre, Nancy. 1994. Who Pays for the Kids? Gender and Structures of Constraint, New York and London: Routledge. [Crossref] , [Google Scholar]) from a gender perspective, and George Akerlof and Rachel Kranton (2000 Akerlof, George and Kranton, Rachel. 2000. “Economics and Identity.”. Quarterly Journal of Economics, 115(3): 715–753. [Crossref], [Web of Science ®] , [Google Scholar]). 3How children absorb gender roles continues to be debated. Social learning theory (Julian Rotter 1982 Rotter, Julian B. 1982. “The Development and Application of Social Learning Theory”. New York: Praeger. [Google Scholar]) emphasizes the importance of direct reinforcement and modeling in shaping children's behavior and attitudes. Cognitive theories, such as gender schema theory, posit that children very early recognize that they are a boy or a girl, not both (Sandra Bem 1981 Bem, Sandra L. 1981. “Gender Schema Theory: A Cognitive Account of Sex Typing.”. Psychological Review, 88: 354–364. [Crossref], [Web of Science ®] , [Google Scholar]). This categorization serves as a magnet for new information and the child begins assimilating new experiences into this schema. Broad distinctions between what kinds of behaviors and activities go with each gender are acquired by observing other children and through the reinforcement they receive from their parents. 4It should be noted that while Iversen and Rosenbluth (2005 Iversen, Torben and Rosenbluth, Frances. 2005. “Gender Socialization: How Bargaining Power Shapes Social Norms and Political Attitudes”. Department of Government, Harvard University and Department of Political Science, Yale University. Working Paper [Google Scholar]) may be accurate in their representation of agricultural societies in earlier periods, their generalization of the gender division of labor is not universally applicable today. For example, in Sub-Saharan Africa, both women and men participate in agriculture, but often there is a gender division of labor in crop production. 5“Distress” sales of labor refer to women's increase in paid labor time in response to declines in income or wages of other family members (often husbands), with the idea that in order to maintain a target level of household income, women must increase their time in paid labor. 6Structural adjustment programs that lead to economic crisis, informalization of labor, and more insecure work conditions have also, in some cases, created the conditions for an increase in sex work – one of the few viable alternatives for generating income for women (Kamala Kempadoo 1999 Kempadoo, Kamala. 1999. Sun, Sex and Gold: Tourism and Sex Work in the Caribbean, Lanham, MD: Rowman and Littlefield. [Google Scholar], 2004 Kempadoo, Kamala. 2004. Sexing the Caribbean: Gender, Race, and Sexual Labor, New York and London: Routledge. [Crossref] , [Google Scholar]). Both women and men (but primarily women) are engaging in this work. Given the sexualization of women's paid labor, it is unclear to what extent this affects gender norms. 7Post-industrial societies are defined in this study as the twenty most affluent countries in the world. 8Morrisson and Jütting (2005 Morrisson, Christian and Jütting, Johannes. 2005. “Women's Discrimination in Developing Countries: A New Data Set for Better Policies.”. World Development, 33(7): 1065–1081. [Crossref], [Web of Science ®] , [Google Scholar]) argue that these institutions are exogenous in that they have been in practice for many years, if not decades or centuries. For this reason, they view the customs as pre-dating current trends in female employment. 9It is useful at this point to recall the evidence on the U-shaped relationship between female labor force participation and GDP per capita. This finding implies that poor and rich countries experience high female labor force participation rates, whereas middle income countries are characterized by low female labor force participation. Theorists have attributed this relationship to changes in labor market structure, social norms regarding the nature of women's work, and cultural factors such as religion, social mobility, and family structure (see, for example, Claudia Goldin 1995 Goldin, Claudia. 1995. “The U-shaped Female Labor Force Function in Economic Development and Economic History,”. In Investment in Women's Human Capital, Edited by: Paul Schultz, T. 61–90. Chicago: University of Chicago Press. [Google Scholar]). This argument is compatible with that made by Inglehart and Norris (2003 Inglehart, Ronald and Norris, Pippa. 2003. Rising Tide: Gender Equality and Cultural Change Around the World, Cambridge, , UK: Cambridge University Press. [Crossref] , [Google Scholar]) who assert that modernization of cultural norms leads to gender equitable changes in norms and stereotypes. Using a relative measure, Cagatay and Olzer (1995 Cagatay, Nilufer and Ozler, Sule. 1995. “Feminization of the Labour Force: The Effects of Long-Term Development and Structural Adjustment.”. World Development, 23(11): 1883–1984. [Crossref], [Web of Science ®] , [Google Scholar]) find that at earlier stages of development, women's share of the labor force falls and attribute this to urbanization and a separation of productive from reproductive work, with women finding it difficult to combine both roles. As growth proceeds, however, relative female labor force participation rises with the commodification of domestic labor, falling fertility, and more education for women. Both approaches use per capita income as a measure of the stage of development. This suggests that in the empirical analysis, there may be some collinearity between GDP and women's share of employment or feminization of the labor force. 10Two possible sources are the International Labour Organization's Yearbook of Labour Statistics (various years) and the United Nation Development Programme's (UNDP) Human Development Report (various years). In the former, gender-disaggregated wage data, primarily for the manufacturing sector, are reported, but data for many countries are missing. With regard to the UNDP, which provides data on female share of income, data are often only estimated, based on the assumption that on average women earn 75 percent of men's income. This is because, again, country-level data that can be used to make reliable estimates are simply not available. I have, therefore, used only female access to paid employment as a second best alternative. 11Psychologists who investigate the dynamics of gender norms and stereotypes, however, are not so much focused on the length of time it takes for change to occur as they are on the factors that induce change. 12Attitudes, of course, are not strictly speaking norms and stereotypes. But attitudes towards various subject matters are based on the underlying set of gender definitions that a person holds. 13For a more detailed discussion of the World Values Survey's sampling methods, see Inglehart, Basáñez, Díez-Medrano, Halman, and Luijkx (2004 Inglehart, Ronald, Basáñez, Miguel, Díez-Medrano, Jaime, Halman, Loek and Luijkx, Ruud. 2004. Human Beliefs and Values: A Cross Cultural Sourcebook Based on the 1999 – 2002 Surveys, Mexico City: Siglo XXI Editores. [Google Scholar]). The Kish-grid system ensures that the household member to be interviewed is selected entirely at random and has an equal chance of being interviewed. It thus avoids the possible bias that can be caused by interviewers interviewing only the most accessible household members. The modified quota sampling approach used is described as follows. “Some surveys used a probability model (area sampling) down to the household level, but switched to a quota design at this last stage” (2004: 390). 14No countries from the Middle East and North Africa (MENA) were included in the 1990 wave. 15The gap is positive in those cases where the question is posed in a gender equitable fashion and negative when the question is posed such that agreement suggests a more patriarchal attitude. 16To some extent, an explanation for the declines in the transition economies must be found at the country level. The data reveal a wide dispersion in average male and female attitudes to the question about jobs. For example, in Russia the percentage of men who agreed with this statement was 50.0 percent in 1990, falling to 42.5 percent in 2000, while Slovenia reveals, on average, more equitable norms held by men, with 34.0 percent agreeing in 1990 and 17.9 percent in 2000. 17Regionally disaggregated data, however, show a sharp decline in feelings of control over lives in Sub-Saharan Africa. 18It is sometimes claimed that women are more altruistic than men, that is, that women exhibit greater empathy (James Andreoni and Lise Vesterlund 2001 Andreoni, James and Vesterlund, Lise. 2001. “Which is the Fair Sex? Gender Differences in Altruism.”. Quarterly Journal of Economics, 116(1): 293–312. [Crossref], [Web of Science ®] , [Google Scholar]; Alessandro Innocenti and Maria Grazia Pazienza 2006 Iversen, Torben and Rosenbluth, Frances. 2005. “Gender Socialization: How Bargaining Power Shapes Social Norms and Political Attitudes”. Department of Government, Harvard University and Department of Political Science, Yale University. Working Paper [Google Scholar]). This may also be a factor motivating women's greater support for redistributive programs (Torben Iversen and Frances Rosenbluth 2006 Iversen, Torben and Rosenbluth, Frances. 2006. “The Political Economy of Gender: Explaining Cross-National Variation in the Gender Division of Labor and the Gender Voting Gap.”. American Journal of Political Science, 50(1): 1–10. [Crossref], [Web of Science ®] , [Google Scholar]; Torben Iversen, Frances Rosenbluth, and David Soskice 2005 Iversen, Torben, Rosenbluth, Frances and Soskice, David. 2005. “Divorce and the Gender Division of Labor in Comparative Perspective.”. International Studies in Gender, State and Society, 12(2): 216–242. [Google Scholar]). Insofar as this is viewed as a fixed trait, we would not expect to see evidence that gender differences in empathy change over time. And yet, the evidence presented in this paper suggests changes over time in both men's and women's attitudes towards redistribution. 19GLS is a method for dealing with cross-sectional heteroskedasticity. Heteroskedasticity may be a problem in a panel data analysis such as this if, for example, the quality of enumeration differs between countries, leading to larger variances in responses. I also conducted these analyses with OLS and results were very similar. 20GDP is inherently a problematic measure. While it is often viewed as an indicator of access to material resources, it undercounts much of women's production and gives credit to some kinds of economic activity that have harmful effects. There is a good deal of research that shows that growth is not equal to well-being, and indeed, it was precisely this recognition that led to the development of Amartya Sen's capabilities approach (Sen 1999 Sen, Amartya. 1999. Development as Freedom, New York: Anchor Books. [Google Scholar]). Interpretation of the effects of this variable then should be viewed with caution and a clear understanding of the limitations of this measure. In fact, it may be useful to read this variable as a measure of commodification of economies rather than as an indicator of the size of the economic pie. The question of how to measure material resources available for distribution is one that feminist economists have been grappling with in recent years when exploring the impact of globalization-cum-liberalization on well-being. Absent a more generalizable measure of material well-being, however, GDP remains the single quantifiable measure widely available. 21Thus, for example, the average of female share of employment from 1985 – 9 is used to explain gender norms and stereotypes in 1990 and so on. 22Agriculture's share of GDP in 1990 is used to explain gender norms in stereotypes in 1990. For consistency, it would have been useful to measure agriculture with the same lagged approach as used for the remaining variables. However, that would have caused a dramatic reduction in the sample size since, for most of the transition economies, these data are not available before 1990. 23Robustness checks seek to determine how sensitive the results are to the model specification. In this case, I vary the choice of variables used to measure female access to work and the structure of the economy to assess whether these variables continue to produce the same effect on norms and stereotypes. 24For the regressions (not reported here) in which I did use female relative wages, that variable was insignificant. But the sharply reduced sample size leads me to be cautious about those results. 25Inglehart and Norris (2003 Inglehart, Ronald and Norris, Pippa. 2003. Rising Tide: Gender Equality and Cultural Change Around the World, Cambridge, , UK: Cambridge University Press. [Crossref] , [Google Scholar]), as noted, also make the point that the relationship between the level of development and gender equitable attitudes may well have to do with the role of the state in implementing policies such as affirmative action, equal pay, reproductive rights, and equitable access to education may play an important mediating role. Governments in higher income countries have been more active in implementing such policies, although certainly some lower income countries have also adopted some of these policies (e.g., Viet Nam and a number of transition economies). Thus, level of development does not adequately proxy for the role of the state in influencing gender norms and stereotypes. 26In the regionally disaggregated data, we find that in the OECD and LAC regions, a higher percentage of women than men agreed with this statement.