103087 Bolivia: Challenges and Constraints to Gender Equality and Women’s Empowerment A © 2015 International Bank for Reconstruction and Development/The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/ igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Bolivia: Challenges and Constraints to Gender Equality and Women’s Empowerment. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Photos: ©Jonna Lundvall, iStock® by Getty Images and Shutterstock Inc.-Bigstock. Bolivia: Challenges and Constraints to Gender Equality and Women’s Empowerment Poverty and Equity Global Practice (GPVDR) June 2015 i ii Table of Contents Acknowledgments................................................................................. iv Objective of the Note............................................................................. 1 Framing the Issues................................................................................... 3 Scope and Limitations 5 1. Endowments........................................................................................ 6 Important Progress towards Gender Parity in Education, with Disparities among Indigenous Groups Remaining 6 Improving Maternal Health: Still a Key Challenge 10 2. Economic Opportunities................................................................... 15 Participating in the Labor Force, but on Unequal Terms 15 Entrepreneurship: Concentrated in Small, Informal Enterprises 21 3. Agency............................................................................................... 23 High Levels of Female Participation in Social and Political Organizations 24 Young and Married Women at a Particular Risk of Violence 26 Teenage Pregnancy Dropping, but Still Above Regional Averages 28 Summary of Findings and Further Research........................................ 31 Annex 1. Methodological Note and Results of the Census Data Analysis......................................................................................... 36 Annex 2. Methodological Note on the Perception Data Analysis...... 39 References............................................................................................. 40 iii Acknowledgments This Note was prepared as part of the Bolivia Multidimensional Poverty Non-Lending Technical Assistance (P132327-P151095) led by the Poverty and Equity Global Practice of the World Bank. The work was made possible thanks to the support of two multi-do- nor trust funds: the Umbrella Facility for Gender Equality (UFGE) and the Rapid Social Response Program (RSR). The UFGE is designed to strengthen awareness, knowl- edge, and capacity for gender-informed policy making; and the RSR provide support to help the world’s poorest countries build effective social protection systems. The analysis and writing of this Note was led by Jonna Lundwall, with the following team members: Santiago Garriga, Anna Tabitha Bonfert, Emcet Tas, and Maria Cecilia Villegas. The Note benefited from the comments of Erwin Tiongson, Maria Ana Lugo, Jennifer Solotaroff and Maria Eugenia Genoni, comments that strengthened the analysis and interpretation. The team wishes to thank the National Institute of Statistics (INE, for its acronym in Spanish) and the women’s rights NGO La Coordinadora de la Mujer for providing access to the data used in this study. Furthermore, the team thanks Karem Edwards Izquierdo, Julio Ricardo Loayza, Veronica Cronembold and Gabriela Orozco, as well as the World Bank Country Office in La Paz for the continuous support to strengthen the dialogue on gender issues in Bolivia. The work was conducted under the general guidance of the Practice Managers Louise Cord and Oscar Calvo-Gonzalez. iv Objective of the Note T o assess the development challenges gender-specific disparities in development out- of Bolivia it is important to consid- comes, highlight opportunities and constraints er gender relations and the situation to women’s empowerment, and identify areas in of women. Although the country has achieved which continuing knowledge gaps are particular- some progress towards gender equity in the past ly important to understand and address gender decade, women still lag behind men in important inequalities. dimensions of well-being. Access to endowments and economic opportunities has improved, with The analysis in this Note suggests that: higher enrollment in school, higher female labor •• There is important progress in develop- force participation, and a higher share of wom- ment outcomes for women, in their ac- en in Parliament and the Cabinet compared to cess to services and to decision-making just 10 years ago. However, outcomes for wom- spaces. en continue to be poor, both in absolute terms •• The intersection of gender and ethnicity and in comparison to men. To name a few, more deepens the gaps in some development women than men are illiterate, the adolescent outcomes in Bolivia. fertility rate is slightly higher than the regional av- •• Women feel discriminated against in dif- erage, and violence against women is among the ferent aspects of their lives, in particular highest in the region. The importance of ethnicity indigenous women. in Bolivian society contributes to the complexity •• Not having access to economic opportuni- of such issues, making some men and women ties limit women’s agency; while women particularly disadvantaged and halting progress with higher levels of agency are able to in multiple areas of development. more fully take advantage of existing opportunities. This Note aims to provide information and analysis as a basis for a better understanding of the challenges and constraints of achiev- ing gender equality in Bolivia, with a special focus on the intersectionality between gender and ethnicity. Combining and analyzing existing evidence and new data, it seeks to document 1 2 Framing the Issues G ender equality has an intrinsic value decline in fertility and a later age of marriage, in itself as a human rights objective. helped increase economic opportunities for If development is a process aimed at women and expand female labor force partic- expanding freedoms equally for all human be- ipation by 15 percent from 2000 to 2010. As ings (Sen, 1999), gender equality constitutes a result, women have played a central role in an important dimension of assuring well-being the dramatic decline of poverty and inequality in for both men and women. However, the reality LAC: gains in female labor income contributed is that gender disparities persist and are often 30 percent of the reduction of extreme poverty manifested in the laws, policies, and practices and 28 percent of the decline in inequality in the of institutions. Furthermore, traditional norms last decade (World Bank, 2012). and practices, including informal justice mech- anisms, may perpetuate violations of the rights The conceptual framework proposed by the of women and girls (United Nations, 2014). 2012 World Development Report on Gender Consequently, strengthening gender equality is Equality and Development is particularly a core development objective in and of itself, helpful in analyzing gender equality. The as it enhances women’s rights and promotes a framework suggests that progress in achieving more equitable society. gender outcomes result from households’ de- cisions, which are influenced by a complex in- Gender equality has instrumental value as teraction across formal and informal institutions well, since it can enhance productivity, im- and markets. Three aspects of equality that prove development outcomes, and make matter in and of themselves, but are also close- institutions more inclusive. Evidence shows ly interlinked, are identified: (1) endowments, that gender disparities can slow down economic (2) economic opportunities, and (3) agency. growth, hamper poverty reduction, and under- mine well-being outcomes for men and women The accumulation of endowments (defined as alike (World Bank, 2011a). The Latin America education, health, and physical assets) has an and the Caribbean (LAC) region, however, has intrinsic value, but also enables individuals to managed to partially defy this trend. Higher use those endowments to benefit from econom- investment in human capital, together with a ic opportunities and generate income (World 3 Bank, 2011a). While outcome disparities in minority women are drastically different from endowments and economic opportunities are those of ethnic majority women, although both well-documented across Latin America, includ- groups fare worse than men in most outcomes. ing Bolivia, agency is the least studied dimen- 1 Intersectionality means that people are simul- sion of gender equality. Agency is a person’s taneously situated in multiple social structures, ability to make choices and to transform them and when these structures intersect, identities into actions. It is not only about having aspira- can produce a multiplication of advantage or tions, but ensuring that the individual does not disadvantages (World Bank, 2013 and Tas, face immovable barriers in exploiting opportu- Reimao, and Orlando, 2014). Existing evidence nities and assets to achieve his or her goals. bears out some of these disadvantages. For Differences between men’s and women’s agen- example, there are sizable gender- and eth- cy usually work to women’s disadvantage and nicity-based wage gaps between indigenous have spillover effects on multiple dimensions of and non-indigenous populations across Latin well-being. These gendered differences matter American countries. These relate to differenc- not only for women, but for their families and es in educational attainment between men and societies as a whole. The agency of women in- women and between indigenous and non-in- fluences their ability to build their human capi- digenous persons, as well as to the scarcity tal and take up economic opportunities, which of indigenous groups in the highest-paid jobs in turn impact their children’s access to health, (Ñopo, Atal, and Winder, 2010). More general- educational, and economic opportunities. ly, indigenous women have been shown to face Empowering women as economic, political, and discrimination both as indigenous people and social actors can change policy choices and as women, resulting in lack of access to edu- make institutions more inclusive. cation, health care, and ancestral land rights. In addition, they face disproportionately high In Bolivia, where about 50 percent of peo- rates of poverty and exposure to domestic and ple identify themselves as indigenous or sexual violence, exacerbated in contexts of traf- Afro-descendants according to the 2012 ficking and conflict (ECOSOC, 2010 and Hall Population and Housing Census, gen- and Patrinos, 2006, 2012). The country’s larg- der-based disparities are largely intertwined est indigenous groups are the Quechua and the with ethnicity. A recent World Bank report Aymara, both considered to be Andean or high- on social inclusion defines intersectionality as land people. In addition, there are approximate- “the understanding that people are simultane- ly 30 officially recognized lowland indigenous ously situated in multiple social structures and people, mainly situated in the Amazon region, realms, which interact in complex ways to influ- the East, Chiquitanía, and Gran Chaco. ence human experiences, social relations, and outcomes” (World Bank, 2013b). The concept of intersecting identities is helpful in explaining Following the WDR2012 framework and ex- the occurrence of additive or multiplied dis- ploring the role of the intersectionality of advantage (or advantage). However, it is also gender and ethnicity, the Note will analyze important to acknowledge the heterogeneity of the situation of Bolivian women in the three indigenous populations. The feminist literature dimensions of gender equality: endow- highlights that day-to-day experiences of ethnic ments, economic opportunities, and agency. 1 See, for example, Aritomi et al. (2010), Chioda (2011), Camacho et al. (2003), Farah et al. (2009), Hunt (2008), Medeiros et al. (2007), Ñopo et al. (2009), Sakho et al. (2009), Silva and Batista (2010), and USAID (2012). 4 Scope and Limitations Bolivia, more analysis and policy discussions will be required, as outlined in the final section In documenting systematic disparities in de- of the Note. Finally, it is important to point out velopment outcomes along various indica- that this document does not provide a compre- tors, the Note’s authors synthesize the em- hensive overview of policies and programs that pirical literature of the last decade with new are already in place to respond to gender and data. In particular, the authors analyze data ethnicity challenges, but instead touches on from the 2011 and 2013 Household Surveys and the legal framework and specific interventions the 2012 Population and Housing Census to or areas of concern, such as conditional cash produce selected indicators. They draw on find- transfers and gender-based violence. ings from a World Bank-supported Perception Survey of Gender Discrimination and Exclusion that was recently conducted in Bolivia to shed light on women’s capacity to take advantage of existing services and economic opportunities. The Note also explores whether the gaps be- tween men, women, and other social groups are manifestations of deeper processes of exclusion that block positive development outcomes for certain members of Bolivian society. Given that about half of all Bolivians identify themselves as indigenous, specific attention has gone to an- alyzing well-being outcomes along ethnic divi- sions. In the Note, the term “indigenous” refers to all people self-identifying as members of an indigenous or native group, as well as Afro- descendants.2 Consequently, the term “non-in- digenous” refers to all people who did not iden- tify themselves as members of these groups. Annex 1 provides additional methodological information on the census data analysis; Annex 2 explains use of the Perception Survey data. While this Note constitutes an important first step towards a better understanding of gen- der-specific challenges and constraints in 2 See Annex 1 for more details on the identification of the ethnicity variable. Alternative indicators of ethnicity have been proposed, such as mother tongue, either combined with self-identification or in isolation. However, to maintain a consistent analysis, this Note uses self-identification as the basis for defining the indigenous and non-indigenous variable, as does the Perception Survey. The use of mother tongue is particularly useful when analyzing data over a longer period of time, due to fluctuations in self- identification, but because this Note focuses on recent data, there is no significant difference between using mother tongue or self-identification to determine the ethnicity variable. 5 1. Endowments I nvestments in education and health shape across ethnic groups and rural versus urban the ability of men and women to reach their residents. Household data from 2013 indicate full potential in society, allowing them to that 15 percent of adult women in Bolivia have take advantage of economic opportunities no schooling, while the corresponding figure for and lead a productive life. Gender differences men is only 4.6 percent. Similarly, 31 percent of in investments in education and health may not women had completed secondary school, com- only adversely affect individual outcomes, but pared to 37 percent of men. A gender gap in lit- may have significant costs for societies at large eracy rates is apparent in the latest 2012 census (World Bank, 2011a). data: in urban areas, 99 percent of men (ages 12 years or older) are literate as opposed to 96 per- cent of women, whereas in rural areas the cor- Important Progress responding statistics are 94 percent versus 84 towards Gender Parity in percent. Moreover, in both urban and rural areas, Education, with Disparities indigenous women face an additional disadvan- among Indigenous Groups tage vis-à-vis all other groups: their literacy rate Remaining was 82 percent in rural areas and 94 percent in urban. Universal education policies in Bolivia date back to the 1930s. A special department Women in rural areas and women who belong was opened at that time within the Ministry of to indigenous groups have lower education Education to tackle low education levels in ru- outcomes than any other group. Literacy and ral and indigenous areas. Further educational school completion rates in rural areas are gen- reforms followed in the 1990s, and in the last de- erally better for men than women, regardless cade additional focus has been put on the right to of indigenous/non-indigenous status. In fact, multicultural and bilingual education. indigenous males fare nearly as well as non-in- digenous males and are slightly better off than Though education is today a universal right non-indigenous females. For example, the litera- in Bolivia, indicators show persistent dis- cy rates for non-indigenous and indigenous men parities between men and women as well as who are 12 years or older are similar (96 and 6 94 percent, respectively), but the Figure 1. Gráfico 1.Completion rates in Tasas de término enurban Bolivia la Bolivia urbana rates for non-indigenous and in- Percentagede Porcentaje male/female population of hombres/mujeres, by indigenous por condición status indígena digenous women in the same age group are further apart (90 and 90 80.8 80 81 percent). Compared to non-in- 71.9 69.7 70 digenous men, moreover, the lit- 60 56.3 48.4 51.3 eracy rate for indigenous women 50 % 40.4 40 is 15 percentage points lower. In 30 28.7 terms of school completion rates 20 10 among individuals aged 25 years 0 or older, indigenous women fare Male Female Male Female Non-indigenous Indigenous even worse. Their primary school Primary school completion Secondary school completion completion rate in rural areas is half the rate for non-indigenous Source: Authors’ calculations based on 2012 Census data. men, at 26 versus 53 percent. The disparity in the secondary school completion rate is also large, at 23 2.Completion Figure 2. Gráfico rates in Tasas de término enrural Bolivia la Bolivia rural Percentagede Porcentaje male/female population of hombres/mujeres, by indigenous por condición status indígena versus 10 percent (see Figures 1 and 2). 90 80 70 In urban areas, education out- 60 52.5 comes are generally higher for 50 41.2 44.5 % 40 all groups compared to rural 30 22.9 25.6 areas, but gender and indige- 20 17.3 18.7 9.8 nous status are still associated 10 0 with lower literacy and school Male Female Male Female Non-indigenous Indigenous completion. While the disad- Primary school completion Secondary school completion vantage of being indigenous is significant in both primary and Source: Authors’ calculations based on 2012 Census data. secondary school completion, it is particularly large in secondary. Only 26 percent of indigenous 3.Literacy Figure 3. Gráfico Tasas derates in urbanen and alfabetismo rural áreas areas y rurales urbanas Percentage of male/female population by indigenous Porcentaje de hombres/mujeres, por condición indígenastatus women 24 years or older report secondary school as their high- 99.1 100 96.0 97.8 98.3 94.0 93.8 est level of education vis-à-vis 34 90.6 81.7 80 percent of non-indigenous wom- en.3 Furthermore, household data 60 % indicates that education levels 40 among Quechua women and men 20 are particularly low: 33 percent of Quechua women and 12 percent 0 Male Female Male Female of Quechua men report having Non-indigenous Indigenous no schooling at all. However, the Rural areas Urban areas Source: Authors’ calculations based on 2012 Census data. Note: World Bank team’s calculation using Census 2012. Primary school com- 3 Authors´ calculation based on the 2013 pletion was defined as completing grades 0-6; secondary school com- Household Survey. pletion defined as completing grades 7-12. 7 to non-indigenous men. In secondary school completion, the gender-based gap is smaller between non-indigenous men and non-indig- enous women (6 percentage points), but the gap is larger between indigenous men and non-indigenous men (11 percentage points). As before, indigenous women face an additional penalty of 6 percentage points, which results in them being 23 percentage points less likely to complete secondary school than non-indige- nous men (see Table A.2 in Annex 1). However, not all indigenous women face the same extent of disadvantage. There are sig- nificant differences in the size of the gender ed- ucation gap across different indigenous groups and age cohorts. The 2012 Census shows that the Aymara population has the largest gender gap in literacy, primary school completion, and secondary school completion. This is despite having the highest overall attainment levels largest gap between women and men in terms among indigenous groups (and second high- of total lack of schooling occurs in the Aymara, est in the country, after non-indigenous groups) among whom 82 percent of the surveyed people and despite residing in urban centers around with no schooling are female. La Paz, where school infrastructure and ed- ucation outcomes are, in general, better than The intersection of gender and indigenous the national average. In contrast, other indig- status has a cumulative negative impact on enous groups have smaller gender gaps, but educational attainment among indigenous their overall attainment levels are also lower women, even after controlling for the impacts than those of Aymaras. For example, second- of age and geographic area. Multivariate anal- ary school completion rates of Aymara men and ysis of the 2012 census data shows that com- Aymara women (ages 14-plus) are 46 percent pared to non-indigenous men 25 years or older, and 32 percent, respectively, with a gender non-indigenous women of that age group are gap of 14 percentage points. In comparison, 9.1 percentage points less likely to complete the corresponding rates for Quechua men and primary school, while indigenous men are 9.7 Quechua women are 30 percent and 25 percent, percentage points less likely to do so.4 The size with a gender gap of only 5 percentage points. of the penalty multiplies when indigenous status This suggests that the greatest disadvantage interacts with gender. In particular, indigenous to Quechua women’s educational achievement women face an additional disadvantage of 8.5 stems from their indigenous identity, but Aymara percentage points for being both indigenous women are disadvantaged both by being indige- and women, which translates into a cumulative nous and female. disadvantage of 27 percentage points relative In general, women’s access to education in Bolivia has improved in recent years, espe- 4 Controlling for age, age-squared, age cohorts and rural/ urban residence. See Annex 1 for methodology. cially among younger cohorts. The boy-girl 8 gap in the completion of primary schooling and The persisting (though diminishing) gaps in the female-to-male enrollment ratio for second- educational enrollment and attainment be- ary schooling are close to parity: as of 2013, 90 tween urban and rural children, as well as percent of boys 12 to 17 years old were attend- between indigenous and non-indigenous ing school, as were 91 percent of girls in the groups, are associated with other direct and same age bracket. While there is not a large dif- indirect factors that can raise drop-out rates. ference between the shares of indigenous and As recognized in the literature,6 these include non-indigenous youth attending school, a large early pregnancy or marriage; lack of separate gap remains depending on residence, with 93 school bathrooms for girls, which is particularly percent of the urban population 12 to 17 years problematic as they get older and start to men- old attending school, compared to 87 percent struate; lengthy distances between schools and for this group in the rural population. home, and perceptions that travel is a “risk;” the pull of the labor market and care responsibilities; Cumulative disadvantages faced by indige- and girls’ higher rates of being victims of abuse.7 nous women have become smaller over time. In more indirect terms, in households that have limited resources, boys are often educated at Multivariate analysis shows that, in comparison the expense of girls and gendered curriculum to non-indigenous men, Aymara women ages and schooling practices tend to silently exclude. 20-29 are only 17 percentage points less like- There is more pressure for girls to leave school ly to complete secondary school, compared to as they get older, and thus gendered patterns 39 percent among Aymara women aged 50-59. can be accentuated at the secondary level (Hunt, Similarly, for Quechua women in these two age 2008). The presence of younger siblings of pre- groups, secondary school completion is estimat- school age at home increases the probabilities of ed to be 20 percentage points (ages 20-29) and older children being out of school and in domes- 34 percentage points (ages 50-59) lower than tic activities such as caring for them, especially that of non-indigenous men. Non-indigenous for rural and indigenous girls (Zapata, 2011 and women, on the other hand, face the smallest Yáñez et al., 2011). cumulative disadvantage, with non-indigenous women in these two age groups being only 0.06 The Perception Survey confirms that in and 6.1 percentage points less likely to complete Bolivia, domestic care work, pregnancy, and secondary school than non-indigenous men, re- the pull of the labor market are common fac- spectively (see Table A.5 in Annex 1). These find- tors impeding girls from attaining higher lev- ings imply that older women in all ethnic groups els of schooling. Notably, 44 percent of female are much less likely to complete primary or sec- students who discontinued their studies say ondary school compared to non-indigenous men that limited financial resources forced them out and non-indigenous women, but the magnitude of school, with similar rates for indigenous and of the disadvantage is the largest for Aymara non-indigenous women (see Figure 4). The lack women in all but the youngest age cohorts (see of economic resources appears to be a more Tables A.5 and A.6 in Annex 1).5 binding constraint in urban settings: 45 percent of women residing there singled it out, compared to 42 percent among their rural counterparts. 5 For literacy, it is Quechua women aged 20-59 and women aged 14-19 who belong to other indigenous groups who face the largest cumulative disadvantage. Aymara women have the largest disadvantage only in ages 60 and above. However, it is important to note that the differences in young 6 For example, see the 2012 WDR. women’s cumulative literacy disadvantages are small across different ethnic groups. 7 For example, see Zapata (2011). 9 groups recall fewer incidents 4.5. Gráfico Figure Discriminación Barriers en el sistema educativo to education Porcentaje de mujeres que han sufrido discriminación, por grupo of discrimination because they Percentage of women who dropped out of school because of*: de edad have been out of the educa- tion system for a while. An al- 11.0 Pregnancy 9.4 ternative hypothesis points to Domestic and 11.1 increased awareness of the care work 9.6 issue of discrimination among Joined the labor market 13.5 younger cohorts and thus 9.3 higher reported incidence.8 Lack of economic 43.6 resources 44.7 These barriers may contribute 0 5 10 15 20 25 30 35 40 45 50 to lowering the overall level of % schooling: almost all Bolivian Non-indigenous Indigenous women (92 percent) say they * Question asked open-endedly and up to two constraints were recorded would have wanted to study Source: Authors’ calculation based on Perception Survey, 2014. more. Another barrier to education is the wide- Improving Maternal Health: spread discrimination that girls and wom- Still a Key Challenge en face in the Bolivian education system. According to the Perception Survey, an average Maternal health is a serious problem in of one in five female students aged 15 to 24 re- Bolivia, which has the highest mortality rates ported having felt discriminated against in ac- in Latin America after Haiti at 310 per 100,000 ademic environments. This aggregate number, live births in 2008, compared to a regional av- however, masks important differences between erage of 80 in 2010. The percentage of births indigenous and non-indigenous women (see attended by skilled staff (71 percent in 2008) is Figure 5): 25 percent of indigenous women com- well below the regional average (90 percent in pared to 18 percent of non-indigenous women. 2010) and has not improved much since 2000’s Interestingly, both the overall incidence of dis- figure of 69 percent (WDI, 2014). In 2013, while crimination and the gap between ethnic groups 97 percent of urban births took place in a health decrease for older cohorts. Arguably, these age institution, only 66 percent of rural births did. The country’s contraceptive preva- Figure 5.5. Gráfico Discrimination Discriminación the inen eleducation system sistema educativo lence rate (61 percent in 2008) Percentage Porcentaje of de women mujereswho havesufrido experienced que han discrimination, discriminación, por grupo is significantly above the world deage by bracket edad average (48 percent). While Bolivia has made progress in 12.0 Older than 41 11.7 terms of the adolescent fertility rate, from 85 births per 1,000 Between 25 and 40 12.9 women ages 15 to 19 in 2000 14.7 18.1 8 For example, see Spears, Brown, Between 15 and 24 and Bigler (2005), who review 25.1 literature and present a model to 0 5 10 15 20 25 30 identify children’s developmental and individual differences that are % likely to influence judgments about Non-indigenous Indigenous discrimination, as well as situational variables that are likely to support Source: Authors’ calculation based on Perception Survey, 2014. these judgments. 10 Box 1. The Bono Juancito Pinto Conditional Cash Transfer Program’s Limited Effect on Access to Education To address gaps in access to education, the government of Bolivia in 2006 initiated Bono Juancito Pinto (BJP), a conditional cash transfer (CCT) program for primary students. Different evaluations have found that CCT programs in general have the potential to improve access to education and help fight inter- generational poverty through the encouragement of human capital accumulation (Schultz, 2004, and Yáñez et al., 2011). Baird et al. (2013) show that while CCTs may have positive effects on enrollment and attendance, particularly for secondary education, there seems to be limited impact on improving test scores. In addition, CCTs provide incentives for families to acquire legal documents such as ID cards for the exercise of political and economic rights (Molyneux and Thomson, 2011). However, whether or not these programs increase school enrollment and completion among the most excluded groups is not as clear. In the case of Bolivia, they have uncertain impact on indigenous groups in general and on indigenous women in particular. Unlike CCT programs for education elsewhere in Latin America, Bolivia’s are universal rather than tar- geted at the poor or groups disadvantaged on the basis of gender or ethnic identity. The Bono Juancito Pinto contributes Bs 200 (almost US$30) per year to every student enrolled in public education up to eighth grade, regardless of their parents’ income status. Students need to have attended at least 80 percent of classes and be less than 18 years old. The BJP payment is given at the end of the school year to the father, mother, or tutor of the student. Household data from 2011 indicates that slightly more men (29 percent) than women (27 percent) collect the Bono and that more people living in urban areas, as well as in the more urbanized de- partments of Santa Cruz, La Paz, and Cochabamba, receive the Bono. This indicates that physical location may affect ability to access the Bono. Since the rural population is less likely to benefit, the program may actually perpetuate rather than alleviate the rural-urban gap in educational outcomes. To the extent that rural residency overlaps with ethnic identity or with specific forms and degrees of gender discrimination in schooling, the in- tervention can also contribute to, rather than redress, disparities based on gender and ethnicity. The data do not indicate any major differences in what women and men use the Bono for, but the largest share of the funding received is spent on clothing and shoes (57 percent), followed by savings (18 percent). Overall, there is little data on how much of the money is spent on actual school-related expenses. Evaluation of the impact of the BJP on enrollment, attendance, and drop-out rates as well as on child labor, poverty, and inequality, has proven difficult. Yáñez et al. (2011) conducted an ex ante impact simula- tion using 2005 data and found that enrollment rates may increase 4 percent as a direct effect of the BJP. Navar- ro (2012) argues that the BJP has had a positive impact on school attendance and a slight impact on preventing drop-out, while its effects on enrollment are not as clear. McGuire (2013) found that, compared to other Latin American CCTs, the BJP was less effective in improving school attendance and enrollment. Two main reasons were identified. The first was that the Bs 200 provided is too little to offset the higher opportunity cost of study- ing. The second reason was related to lack of incentives to send children to school based on the poor quality of education service provision, with a lack of trained teachers, deficient school buildings and infrastructure, and sometimes the need for children to work. In addition, McGuire highlights the fact that the small stipend is unlikely to have a significant impact on income poverty or income inequality, as it amounts to only 4 percent of the average annual consumption of a Bolivian household. In the case of countries where transfers had sig- nificant beneficial effects on enrollment and attendance, the stipends amounted to 27 percent of the average annual consumption for Nicaragua’s Red de Protección Social, 20 percent for Mexico’s Oportunidades, 17 percent for Colombia’s Familias en Acción, and 10 percent for Ecuador’s Bono de Desarrollo Humano. For the BJP, the difficulty of evaluating the impact on child labor, poverty, and inequality stems from the fact that the program did not include a randomized control trial or a baseline study.. 11 maternal mortality rates were La Paz, Oruro, and Potosi, all classified as highly indigenous. These jurisdictions also had the lowest levels of insti- tutional deliveries. Indeed, data from the 2013 Household Survey demonstrate that non-indige- nous women, at a higher rate than indigenous women, give birth at a health institution no matter if they live in an urban or rural area (Figure 6), a finding corroborated in the Perception Survey. In addition, Aymara women tend to use health insti- tutions slightly less than Quechua women both in urban and rural settings. Access and Resource Constraints May Impede Access to Quality Health Care to 72 in 2012, this is still slightly higher than the There are important differences in access regional average of 68. (For further discussion to hospitals and public insurance based on of adolescent pregnancy, see the section about gender, ethnicity, and the intersection of the agency.) two. To ensure adequate care for all women and to fulfill women’s constitutional right to safe moth- Historically, the maternal and child health in- erhood with a “vision and intercultural practice”10, dicators of indigenous and rural populations the Ministry of Health has established protocols have performed worse than those of non-in- for motherhood and newborn care that take into digenous and urban populations. Infant mor- account cultural aspects and customs.11 Not sur- tality is almost twice as high in rural areas as prisingly, a larger percentage of rural residents in urban areas. For example, Silva and Batista depend on public hospitals than do urban res- (2010), using data from the Demographic and idents, 68-76 percent in rural areas, compared Health Survey (DHS),9 found that the gaps in in- to 57-61 percent in urban areas (2012 CNPV, fant health between urban and rural areas had INE). There is also a large gap between indige- deepened: from 1994 to 2008, the infant mor- nous and non-indigenous groups, both in access tality rate in Bolivia as a whole declined by 38 to public/private hospital and to public/private percent, but the fall in rural areas was only 29 insurance. For example, only seven percent of percent. The authors also found a gap between indigenous men and women report using a pri- indigenous and non-indigenous populations with vate hospital in rural areas compared to 14 per- respect to maternal mortality. Given the lack of cent of non-indigenous men and women. Similar ethnic-specific mortality rates, the authors clas- gaps exist in urban areas, but a greater share of sified the country’s nine departments as indige- people use private hospitals and public or private nous if more than 60 percent of the people be- insurance in urban areas relative to rural ones. In long to a native group and as non-indigenous addition, compared to non-indigenous men, in- if more than 60 percent do not belong to any digenous women are 5.1 percentage points less native group. The departments with the highest 10 Article 45 of the Bolivian Constitution states that “Women 9 The Demographic and Health Survey is also known as have the right to safe maternity, with an intercultural vision ENDSA, the Spanish abbreviation for Encuesta Nacional de and practice”. Demografia y Salud. In Bolivia, the latest DHS was carried out in 2008. 11 These include a ministerial resolution, No. 0348/2006. 12 points more likely to use a pub- Gráfico 6. Porcentaje de mujeres que dan a luz en una Figure 6. Share of women giving birth at a health institution, lic hospital and 5 percentage institución de salud by location and ethnic identification Por lugar e identificación étnica points less likely to use a pri- vate hospital. Indigenous 59.5 40.5 Rural Giving birth in Bolivia is cost- Non-indigenous 86.7 13.3 ly and access to insurance to pay for it is uneven. According Indigenous 93.8 6.2 to the 2011 Household Survey, Urban Non-indigenous 98.8 1.2 the average cost for giving birth is around Bs 800,13 with a stark 0 10 20 30 40 50 60 70 80 90 100 difference between rural areas Births with skilled health personnel Other (Bs 530) and urban areas (Bs Source: Authors’ calculation based on the Household Survey 2013. 1,110) and between fees paid likely to use public insurance and 3.6 percent- by indigenous women (Bs 690) age points less likely to use private insurance. and non-indigenous women (Bs 1,070). The Differences in access rates between indigenous highest amounts spent on child birth were report- ed in the departments of Oruro, Santa Cruz, and and non-indigenous women may in part be driv- Cochabamba, while the lowest amounts were in en by perceived discrimination in the health care Potosi, Beni, and La Paz. A larger share of indig- system. According to the Perception Survey, enous women (53 percent) has no insurance that 20 percent of indigenous women report having pays for child birth as compared to non-indige- experienced discrimination when seeking care, nous women (39 percent). Access to both public compared to 14 percent among non-indigenous. and private insurance that covers birth costs is This factor might partly explain decisions by in- higher in urban areas (60 percent) than in rural digenous women not to deliver their children in (48 percent). Quechua and Aymara women living health clinics.12 in urban areas are more likely to have some kind of insurance to cover child birth than members of A large portion of the cumulative disadvan- those groups living in rural areas. For non-indig- tage faced by indigenous women is driven by enous women, the urban-rural difference is not their indigenous status rather than their gen- as significant. der (Table 1). Non-indigenous women are more likely to use any kind of hospital compared to To address the high incidence of maternal non-indigenous men (by 2.5 percentage points and infant mortality and the costs associat- and 0.6 percentage points, respectively), once ed with child birth, the government of Bolivia one controls for age (non-linearly) and area of introduced the Bono Juana Azurduy (BJA) residence. Indigenous men are 0.9 percentage program targeted at pregnant women and points more likely than non-indigenous men to new mothers. The BJA provides a total benefit use a public hospital and 5.3 percentage points worth up to US$ 260. There are separate cash less likely to use a private hospital. Indigenous payments for up to four prenatal medical visits; women, on the other hand, are 4.5 percentage for giving birth attended by trained personnel; acquiring a birth certificate for the baby; getting a 12 Qualitative evidence from Peru suggests that perception of a lack of respect for cultural practices and values can act week of post-partum medical monitoring; and for as a sizeable deterrent to women seeking an institutional birth (World Bank, 2011b). Therefore, access to health clinics alone is not sufficient to increase the percentage of 13 Bs is the symbol for the national currency, the Boliviano. institutional births. In May 2015, Bs 100 were worth approximately US$14. 13 Marginal likelihood of using public and private insurance as compared to non- Table 1. indigenous men Public insurance Private insurance Female 0,005*** -0,003*** Indigenous -0,041*** -0,030*** Female* Indigenous -0,015*** -0,003*** Rural -0,139*** -0,073*** Age 0,022*** 0,001*** Obs. 7.129.134 7.129.134 Pseudo R.-sq 0,05 0,04 Source: World Bank team’s calculation based on the 2012 Bolivia National Census of Population and Housing. Notes: (1) Average marginal effects from probit regressions are reported, using insurance as the dependent variable and non-in- digenous and non-Afro-descendant males as the reference group. ***p<0.01, **p<0.05, *p<0.10. (2) Regressions also control for age-squared and age cohorts. The error terms are clustered at the household level. (3) Primary and secondary school completion rates for individuals 25 and older. taking the baby for up to twelve checkups spaced points from 2007 to 2009. But the impact eval- at two month intervals over a two-year period. uation showed that the BJA has brought a rise Several categories of pregnant women and of almost 5 percentage points in the probability new mothers are ineligible for the BJA, includ- that a birth is attended by health personnel and ing those who are covered by other insurance, an increase in the number of comprehensive or who have given birth or had an abortion with- health checks of children (draft UDAPE evalua- in the past three years. The last two conditions tion, 2014). In addition, the evaluation found no seek to avoid creating incentives for women to significant impact on the number of prenatal con- get pregnant again before a stipulated birth inter- trols or improvements in malnutrition rates. One val of three years has elapsed (McGuire, 2013). of the main reasons for a slight decrease in pre- natal visits may be the scarcity of doctors in four of Bolivia’s nine departments. The rise in patient While the share of births attended by skilled demand there appears to have overwhelmed health staff has increased, it is not clear if this the capacity of small health facilities (McGuire, progress is due to the Bono Juana Azurduy.14 2013). Other findings of the impact evaluation Household survey data show that the percent- include an 8.4 percentage point reduction in low age of births in a health institution increased 20 birth weights and an 8 percentage point increase percentage points, from 67 percent in 2009 to 87 in the early detection of pregnancy (draft UDAPE percent in 2013. According to a recent analy- evaluation, 2014). sis by the government, approximately 420,000 pregnant women and 630,000 children benefited from the program from 2009 to the end of 2013 (draft UDAPE evaluation, 2014). This positive trend predates the BJA, with the share of births attended by skilled staff rising by 4.1 percentage 14 The government’s think tank UDAPE recently concluded an evaluation of the Bono Juana Azurduy, but the results are not yet publicly available. 14 2. Economic Opportunities A ccess to economic opportunities is a force participation during the last decade, while powerful tool to help women escape the male figure has remained more or less static poverty traps and has the potential (Figure 7). Indigenous women have a higher par- to contribute to Bolivia’s overall efforts to ticipation rate (62 percent) than non-indigenous reduce poverty. The female labor force par- women (55 percent).15 Rural areas in general have ticipation rate in Bolivia is one of the highest in higher rates compared to urban areas (80 and 65 Latin America, and the annual growth rate of percent, respectively), and indigenous women in female-to-male labor force participation is also rural areas have a significantly higher rate com- high. However, because women continue to also pared to non-indigenous women—for example, hold primary responsibility for household chores 86 percent among rural Aymara women versus and child-rearing, extensive work hours and in- 61 percent among non-indigenous rural women. formality may mean substantial wellbeing costs for women. Also, despite the large participation The female labor force is characterized by of women in the labor market, they are still gen- vulnerable employment and informality, and erally considered secondary wage earners and by a concentration in low-productivity sec- are paid less than their male counterparts (World tors. Gender inequalities in the labor market are Bank, 2010 and World Bank, 2011a). the product of multiple constraints that accumu- late across the life cycle, not just disadvantages Participating in the Labor at a worker’s productive age. Accumulated con- Force, but on Unequal Terms straints include lack of mobility, time, and skills; exposure to violence; and the absence of basic Bolivian women have a high rate of labor legal rights (World Bank, 2014b). Of total female force participation compared to the region- workers, more than a quarter, 28 percent, are in al average. In 2013, the International Labour part-time jobs. This compares to about a tenth for Organization estimated that the female labor male, 11 percent (data from 2009, WDI, 2014). force participation rate in Bolivia was 64 percent, For 2009, an estimated 64 percent of employed considerably higher than the LAC average of 54 Bolivian women were working in conditions of percent (WDI, 2015). Both LAC and Bolivia have experienced fairly rapid growth in female labor 15 Household Survey 2013 (INE). 15 the 2012 World Development Gráfico Figure 7.7. de participacion Tasa force Labor la fuerza laboral enrate participation Report, Bolivia has the world’s Por lugar e identificación étnica Percentage of female/male population ages 15+ third-largest difference in fe- male-to-male participation in 48.1 LAC 51.4 the informal sector, with only 53.6 Ethiopia and Nepal showing a Female 59.7 larger difference (World Bank, Bolivia 61.5 2011a). The need for flexibil- 80.7 ity, the limited opportunities in LAC 80.1 79.6 the formal labor market, and Male the gender wage gap combine 81.7 Bolivia 81.4 to force women into the infor- 80.9 mal economy. This frequent- 0 10 20 30 40 50 60 70 80 90 100 ly comes at the cost of labor % 2000 2006 2013 rights, pensions, and other benefits (World Bank, 2009 Source: World Development Indicators, World Bank, 2014. and World Bank, 2010). vulnerability,16 more than twice the average of the LAC rate for women, 31 percent. Vulnerable Employed Bolivian women are over-repre- employment has high income variability and may sented in low-productivity sectors. Female end abruptly, destabilizing the social and eco- workers are concentrated in commerce, educa- nomic domains of workers. This type of work of- tion, and health and as domestic workers, while ten undermines the ability of families to borrow male workers are over-represented in the indus- money, repay debt, and manage externalities try, construction, utilities, and transportation sec- and shocks. In addition, workers may have only tors (data from 2008; SEDLAC, 2014). Bolivian limited access to benefits such as pensions. women also feel they have slightly less freedom than men to select their occupation: 66 percent Bolivia has one of the largest informal econ- of men perceive that they are fully or fairly well omies in Latin America, with a considerably guaranteed the freedom to choose their pro- higher share of women working in it than fession, but only 62 percent of women believe men. Sixty-eight percent of female workers are this to be true for themselves (Latinobarómetro, in informal employment, compared to 58 per- 2011). A study of gender in Bolivian production cent of male workers (Figure 8).17 According to suggests that while the concentration of women in certain sectors is the result of preferences, so- cial expectations, or gender stereotypes, gender 16 The World Bank Development Data Portal (2012) defines vulnerable employment as “unpaid family workers and own- gaps within sectors are related to women’s lower account workers as a percentage of total employment.” It says that “own-account workers are workers who, working education and concentration in lower-skill jobs on their own account or with one or more partners, hold within the sector (World Bank, 2009). The level the types of jobs defined as ‘self-employment jobs’ and have not engaged on a continuous basis any employees to of education plays an important role in explain- work for them. Own-account workers are a subcategory of ing gender wage gaps, especially for low-income ‘self-employed.’” 17 Authors’ own estimates based on SEDLAC data workers. A regional study by Ñopo et al. (2009) (CEDLAS and World Bank). The definition of informality is shows that differences in educational attainment based on productivity. Not informal includes an employer or salaried worker in a large or public firm or self-employed explain more about differences in wages among with secondary education completed; Informal includes low-income workers than among middle- or a salaried worker in a small firm, self-employed with incomplete secondary education and non-salaried workers. high-income workers. The study estimates that 16 main significant gaps in opportuni- Figure Gráfico Female 8.7. Tasa deemployment participacion in en LACla fuerza laboral Female population ages 18 to 56 in formal and informal ties to get a job and receive an ad- Por lugar e identificación étnica employment, 2013 equate salary: 31 percent believe that it is easier for a man to get a 80 job than a woman, and 40 percent 70 consider it more difficult for women 60 25 50 24 20 to receive adequate pay. A slightly 42 23 19 higher fraction of indigenous wom- % 40 32 32 15 32 28 30 48 en find it difficult to get adequately 38 43 20 33 35 compensated, while a higher per- 26 31 19 20 23 24 10 centage of non-indigenous wom- 0 hil e ica t ina u ay az il ala do r bia u ay l ivi a er u en expressed the view that in their C ta R en ug Br tem ua lom ag Bo P os Ar g Ur a Ec Co Pa r C Gu experience getting a job is more Informal Formal difficult for women than for men (Figure 10). Perceptions of gen- Authors’ calculation based on SEDLAC data (CEDLAS and the World Source: Bank). der gaps in access to credit and land tenure seem to be less pro- if women had the same employment character- nounced, though a larger share of istics as men in terms of age and education, the non-indigenous women agree that owning land average gender wage gap would almost double. and accessing credit is easier for men. Social norms, in combination with the level of Discrimination in the labor market is a wide- individual agency, can either help or hinder spread problem that may lower women’s ac- women’s capacity to take advantage of exist- cess to economic opportunities. According to ing opportunities, especially in the economic the Perception Survey, almost a quarter of wom- sphere. The Perception Survey finds that a fifth en (24 percent) can recall facing discriminative of women strongly agree that the man should incidents while searching for employment or on- provide for the family, and more than half agree the-job. A higher share of women in urban areas that the man should earn more than the woman. experienced it (28 percent) than those living in This underlines the persistence of gender ste- rural areas (19 percent). Indigenous women reotypes that depict the man as the principal breadwinner. Figure Gráfico Social 9.6. normsde Porcentaje and economic queactivity mujeres dan a luz en una These perceptions do not dif- institución Percentage de of salud who agree with the following women fer much between indigenous statements Por lugar e identificación étnica and non-indigenous women “The woman should provide 30.0 (Figure 9). In general, Bolivian for the family” 30.0 women perceive that they have “The woman should earn more 41.8 money than the man” fewer opportunities to get a 41.2 job than men have, and this “The man should earn more 48.7 money than the woman” 50.9 perception is particularly high “The man should provide 62.5 for women with lower levels of for the family” 62.6 education (Latinobarómetro, 0 10 20 30 40 50 60 70 80 2011). Similarly, the Perception % Non-indigenous Indigenous Survey finds that many women Source: Authors’ calculation based on Perception Survey, 2014. share the opinion that there re- 17 ed women was slightly higher Figure Gráfico 6. Perceived 10. Porcentaje economic de mujeresopportunities que dan a luz en una institución Percentage de of salud who find it more difficult for women women than men’s, while the opposite than for men Por lugar to: e identificación étnica was true (and quite substan- tially) among tertiary-educated 39.0 Get adequate pay 41.3 workers. Given that indigenous 38.6 women are less likely to com- Get a job 35.9 plete primary and secondary Own land 26.7 school than non-indigenous 21.4 men and non-indigenous wom- Access to credit 20.7 en, the relatively poorer labor 16.2 market outcomes for indigenous 0 10 20 30 40 50 % women (such as employability Non-indigenous Indígenas and earnings) may be a man- Source: Authors’ calculation based on Perception Survey, 2014. ifestation of the lasting effects of the poor access to education were particularly affected, with 27 percent having that they had earlier in their lives. experienced discrimination compared to 23 per- cent of non-indigenous women. These internal- Wage gaps remain a challenge in Bolivia. ized social expectations may deter women from Calculating the gender wage is sensitive to factors moving into higher-paying jobs. Research shows such as the methodology used, underlying data, that perceived discrimination can alter both the and sample size. The data are not always compa- expectations that job seekers have of the labor rable across countries. While gender wage gaps market and their decisions on whether to enter it against women are common in Latin America, (Goldsmith et al., 2004 and Das, 2013). they tend to be more pronounced in Bolivia (World Bank, 2009). The gross gender gap in monthly la- In Bolivia, women are more likely than men bor income for the country as a whole increased to be unemployed. The female-to-male unem- from 0.71 in 2000 to 0.65 in 2008 (World Bank, ployment ratio was 1.60 in 2013, and female ad- 2012). This is partly due to wages in female-dom- olescents were more likely to be unemployed than their male counterparts: for ages 15 to 24, Gráfico Figure 11. Tasa de participacion 7.Unemployment enof by level fuerza laboral la educational attainment identificación Por lugar eof Percentage étnica all female/male unemployment 7.6 percent of the female labor force was unemployed in 2011, compared to 6.4 percent of the 70 62 59 young males. There has been 60 an impressive drop in unem- 50 47 47 44 ployment among primary-edu- 40 40 37 33 34 % cated men and women, while 30 26 31 28 24 the share of unemployed people 20 20 22 17 with secondary and tertiary ed- 10 6 ucation has increased (Figure 3 0 11), reversing the standard of Female Male Female Male Female Male higher unemployment rates for With primary education With secondary education With tertiary education 2000 2005 2009 the least educated. In 2009, the rate of unemployment among primary- and secondary-educat- Source: Authors’ calculation based on data from WDI, 2015. 18 inated sectors and occupations tending to be low- er than those in male-dominated sectors and oc- cupations, which limits women’s upward mobility and has important costs in terms of efficiency and poverty reduction. Using qualitative variables con- tained in an Executive Opinion Survey, the World Economic Forum (2013) calculates perceived wage equality for similar work indicators. Bolivia shows a 0.53 female-to-male-ratio, which puts the country among those with the largest perceived gaps, at a ranking of 118 out of 130. In the region, only Uruguay and Chile had a larger gap. Educational and demographic differences between indigenous and non-indigenous groups appear to play an important role in This is corroborated by a recent study by Canelas wage gaps. Indigenous women in Bolivia are more likely than non-indigenous women to be in and Salazar (2014)19 which uses data from 2001 the informal sector and self-employed. More than to examine gender and ethnic wage gaps and 60 percent of indigenous women are self-em- the allocation of time to paid and unpaid work. ployed in the informal sector compared to 40 This study observes that the average wage gap percent of non-indigenous women, with average for indigenous workers of both genders is larg- earnings for indigenous women at 60 percent of er than the average wage gap for woman work- the average for non-indigenous women (World ers in Bolivia. Indigenous workers are paid less Bank, 2009). Furthermore, indigenous popula- than non-indigenous workers, even when they tions have on average four years less of education have the same level of education. Ñopo et al. than non-indigenous ones, which contributed to (2009),20 make a similar finding that ethnicity is the gap in development outcomes and econom- more important than gender in determining the ic participation: the average household income wage gap. They calculate a 31 percent21 differ- per capita is about 60 percent lower among in- ence in the average wage between indigenous digenous households (Zapata, 2011). There are and non-indigenous wage earners, attributed to mixed results from studies that explore the ex- each group’s observable characteristics distinct tent to which gender and ethnic wage gaps can be attributed to differences in observable charac- 19 Canelas and Salazar (2014) study Bolivia, Ecuador, and teristics. The Note team could not find new work Guatemala. They apply the Blinder-Oaxaca decomposition in combination with a quantile decomposition method to on the gender wage gap in Bolivia, and existing compute the effect of each covariate on the unconditional work is generally based on older data. Contreras wage distribution. and Galvan (2003) argue that the intersectional- 20 Ñopo et al. (2009) apply an econometric procedure which is an extension of the Blinder-Oaxaca decomposition ity of gender and ethnicity is the most unfavor- using a non-parametric matching approach. They consider able condition in wage employment in Bolivia.18 five sets of observable demographic characteristics as control variables, added in an order that considers first those that are likely to be less endogenous to a model of wage determination: age, education, presence of children 18 This study uses data from surveys from 1994 and 1999 12 years old or younger in the household (dummy), by the National Institute of Statistics. Based on these presence of other wage earner in the household (dummy), statistics, the study develops income functions aimed at and urban area (dummy). They look at wage earners quantifying the impact of discrimination and its evolution between 18 and 65 years old, based on household surveys over time. To explain income inequalities, the study uses circa 2005. Fields decomposition’s methodology which quantifies every variable from the income equation. 21 As a percentage of average indigenous wages. 19 from the other group. Further, they find that 36 paid work,22 found that 34 percent of the to- percent of the wage gap is unexplained even tal inequality in time spent in unpaid work was after controlling for gender and age, but the due to net differences between women’s and unexplained component falls to 17 percent men’s allocations of time. The differentiation in after controlling for gender, age, and educa- unpaid work occurs mostly in the duration of tional attainment (primary, secondary, tertia- shifts: on average, women’s unpaid work shifts ry, or no education) and to 12 percent after are more than three times longer than those of controlling for gender, age, educational at- men (Medeiros et al., 2007). tainment, and household demographic char- acteristics (urban residence, presence of chil- Bolivian women would spend more time dren in household, and presence of a second working or studying if they would devote income earner). less time to household and care work. The Perception Survey investigates how women There are high levels of inequality in the dis- would spend their time if they were relieved of tribution of paid and unpaid domestic work their domestic and care duties. On average, 43 time among men and women, which per- percent of women would work and 26 percent sistent social norms may partially explain. would study if they had more time at their dis- Men and women allocate their time based on posal. Hence, women’s time commitments re- several constraints, some of them related to sulting from their traditional roles as caretakers social roles based on gender. Despite their and providers of domestic service can prevent increasing role in the labor market, women them from engaging in productive activities. still carry the majority of domestic obligations, There are significant differences by ethnici- which imposes a burden on their time and well- ty (Figure 12): at 50 percent, the proportion of being. This time deprivation affects women’s choices and Figure Gráfico 6. Alternative 12. time Porcentaje de use que dan a luz en una mujeres their personal development. womende institución How salud would spend their time without domestic and Canelas and Salazar (2014) care work e identificación étnica Por lugar show that Bolivian women 39 spend on average four times Work 50 more time performing do- 26 Study mestic tasks than men, and 24 still spend almost 38 hours 18 Rest a week in labor market work 12 (men spend 47 hours). The Personal 4 time 2 two authors also find that lev- 0 5 10 15 20 25 30 35 40 45 50 55 el of education decreases the % number of hours that women Non-indigenous Indigenous spend on domestic work and Source: Authors’ calculation based on Perception Survey, 2014. increases the hours that men spend on these tasks. They further show that members of indigenous pop- indigenous women who would work if they did ulations devote similar numbers of hours to do- not have to attend to domestic or care respon- mestic activities as the rest of the population, sibilities is higher than the equivalent group of but spend about three hours less in paid mar- non-indigenous women (39 percent). ket activities. Another study of Bolivian gender inequalities in allocating time to paid and un- 22 Unpaid work refers to domestic or household activities. 20 Entrepreneurship: Concentrated in Small, Informal Enterprises Starting your own firm is a way to generate in- come when labor market access and options are limited. Many of the reasons why women start and maintain small-sized firms are direct- ly linked to specific gender roles and norms. Female entrepreneurship in Latin America tends to be influenced by women’s position as second- ary income earners. Women still carry the main household and child-caring responsibilities, and smaller enterprises offer more flexibility, though at the cost of growth and earnings (World Bank, 2010b). Married women, for instance, tend to enter the labor market and/or start a formal or informal business to complement household in- comes when bad economic conditions make it necessary for family support. They are not gen- erally seeking to become the household’s single with more than 20 employees (Figure 13).23 An or principal income earner. In contrast, single analysis of 630 Bolivian female-led micro and mothers seek to work regardless of the economic small firms confirms that enterprises led by wom- context because they are primary breadwinners. en tend to be more informal and have lower prof- The highest participation of women in the labor its (World Bank, 2009). The 2010 enterprise sur- market, in fact, is found among single women vey shows that while a significant share of small with no children, with rates close to those of men formally registered companies (34 percent) have (Cunningham, 2001 and World Bank, 2010). a woman as the top manager, female leadership is less common among medium and large firms Female entrepreneurs tend to be more con- (11 percent and 5.7 percent, respectively).24 centrated in low-productive, smaller-sized firms than their male counterparts. Differences In Bolivia, female top managers identify very in productivity are mainly a function of the busi- specific constraints, such as crime and ac- ness size and economic sector, but a range of cess to finance. When women become entre- lesser factors also play a role: differences in lev- preneurs because of constraints in the formal els of education, access to productive resourc- labor market, they face forms of risk and vul- nerability that their male colleagues often do es, networks and markets, and access to train- ing and business development services (World 23 Calculation uses “employer” in household surveys as a Bank, 2011a and World Bank, 2010b). In Bolivia, proxy to ownership, and includes employers between 18 and 65 years old. 84 percent of female-owned firms are microen- 24 The Enterprise Survey identifies firms with five to terprises, with fewer than five employees, 15 19 employees as small; 20 to 99 as medium; and over percent are small enterprises (five to 19 employ- 100 employees as large. For comparison reasons, this document uses this same definition, leaving enterprises with ees), and only 1 percent are medium enterprises one to four employees as micro enterprises. 21 practices of informal sector com- Gráfico Figure 7. Firm 13. Tasa ownership de participacion en by size oflafirm fuerza laboral petitors and political instability Por lugar e Employers identificación between 18 andétnica 65 years old, 2013 as the most problematic issues (Figure 14). Furthermore, the 100 1 3 1 6 1 3 1 2 15 13 2007 Bolivian Micro-Enterprises 20 23 80 27 28 29 30 Survey indicates that 60 percent of female firm owners perceive child 60 and family care responsibilities to % 84 86 40 70 71 65 73 79 68 be an obstacle to firm operation and growth (World Bank, 2010b). 20 0 Female Male Female Male Female Male Female Male Women are less likely than men Bolivia Ecuador Guatemala Peru to use formal credit and financial Micro 1-4 Small 5-19 Medium 20-99 services. Top female managers in Bolivia identify access to finance Authors’ calculation based on SEDLAC data (CEDLAS and the World Source: Bank). as one of the major constraints, but this seems to be demand-driv- not. According to the 2010 Enterprise Survey, en rather than imposed by market entrepreneurs are in general considerably less restrictions. There are no major indicators of gen- worried about corruption, customs, trade regula- der-based discrimination in access to credit and tions, tax administration and rates in Bolivia as financial services, but only 25 percent of women compared to the average for the region (World over 15 years old, compared to 35 percent of men, Bank, 2010a). However, the issues raised as have accounts at a formal financial institution. And main concerns differ very much among women while 7.5 percent of men have accounts used for and men. Women mention the practices of com- business purposes, only 5.5 percent of women do petitors in the informal sector, the lack of skilled (2011 data from WDI, 2014). Gender differences in workers, finance access, and crime, theft, and the use of formal financial services narrow as the disorder as the main constraints; men identify firm grows (World Bank, 2009). Figure Gráfico 7. Top 14. Tasaconstraints experienced de participacion by firms, en la fuerza laboral by gender of top manager Por lugar e identificación étnica Electricity 7.5 Political instability 17.5 3.7 0.4 Tax rates 5.1 Transportation 0.3 5.3 Crime, theft and disorder 1.4 15.5 9.6 Access to finance 16.0 Inaquately educated... 9.4 21.6 Practices of 40.6 informal sector... 23.7 0 10 20 30 40 50 % Male-led Female-led Source: Authors’ calculation based on Enterprise Survey Data, World Bank, 2010. 22 3. Agency I mprovements in agency—a person’s ca- limit the practical impact of the legally recog- pability to make strategic life choices and nized and guaranteed rights of women. The to transform these choices into desired Bolivian legal system guarantees rights such as outcomes—are paramount to speeding up gender parity norms for election to Congress; progress towards gender equality and to de- freedom from violence; protection during preg- velopment in general. While agency in itself nancy; equal pay for work of equal worth; protec- is often hard to measure, some manifestations tion from employment discrimination for reasons of the lack of agency, such as teen pregnancy, of pregnancy, civil status, age, or physical traits; gender-based violence (GBV), and low female reproductive and sexual rights; and equal access political participation, are used as indicators. to land ownership. However, social norms can Understanding agency means understanding limit the effect of laws to the detriment of gender what factors and social norms help or hinder an equality. Extensive literature has demonstrated individual’s capacity to take advantage of exist- that social norms are particularly binding where ing opportunities and assets (World Bank, 2011a an increase in women’s agency would threaten and World Bank, 2014). the balance of power in the household, includ- ing women’s political participation (World Bank, While Bolivia has numerous legislative mea- 2011a). Even where constitutional guarantees sures intended to protect and promote wom- provide for equality and laws protect women’s en’s rights, based on the highly gender-main- rights, discriminatory practices by law enforce- streamed 2009 Constitution, 25 social norms ment and security services, courts, lawyers, and social services can include major obsta- 25 The Constitution enacted in 2009 reflects gender issues cles to women’s security and access to justice. in 23 articles complying with six principles: (1) equity and Customary and traditional norms and practices, positive action, (2) equality and non-discrimination, (3) recognition of specific rights, (4) democratization and including informal justice mechanisms, may per- recognition of the value of domestic work, (5) approval of rights, and (6) non-sexist language–the majority of the petuate gross violations of the rights of women Constitution’s articles apply to both sexes. In addition, and girls (United Nations, 2014). in 2008 the Government launched the National Plan for Equality of Opportunities, including policies in six broad areas: (1) economy, labor and productivity, (2) education, (3) Discrimination is still felt by a large share of health, (4) violence, (5) participation in decision-making, and (6) institutional strengthening. the population. There are many different traits 23 that can be the basis for discrimination, and often and limit their capacities as leaders. Removing they reinforce each other. The Perception Survey the barriers preventing women from taking part not only provides evidence that indigenous wom- in decision-making in the formal arena and influ- en feel more exposed to discrimination, but high- encing the overall policy agenda are key objec- lights the relevance of intersectionality. While tives (World Bank, 2011a). In Bolivia, 46 percent about 13 percent of non-indigenous women in- of women perceive that they are generally or fully dicate that they have felt discrimination on the guaranteed the freedom of political participation, basis of their gender, twice as many indigenous a slightly smaller share than among men, 50 per- women shared that belief. In other words, a more cent (Latinobarómetro, 2011). nuanced examination of drivers of discrimination reveals that overlapping identities (female and Bolivian women are fairly well represented in indigenous) have a multiplier effect and exacer- the political sphere. There are quotas to ensure bate the impact of discrimination (Figure 15). female representation in Parliament—the share of women there increased from 12 percent in 2002 to about 25 percent in 2012 and 49 per- High Levels of Female cent in 2015. For the first time ever, the Cabinet Participation in Social and appointed in January 2010 was comprised of 50 Political Organizations percent women, although by January 2014 this had declined to 33 percent, the same share in Political participation is very important if the Cabinet appointed in early 2015. Also, at the women are to recognize and articulate in- new Tribunal Supreme Court, 33 percent of the terests and build alliances and consensus titular judges are women, as are 77 percent of to advance common projects. Political and the alternate judges, based on the October 2011 electoral systems may impair or facilitate wom- elections. However, the high level of female rep- en’s ability to participate in political processes. resentation at national institutions has not trans- In addition, time constraints and social norms lated into increased participation at sub-national may discourage women’s political participation levels. While the share of female elected officials Gráfico Figure Tasa de participacion en la fuerza laboral 7. Discrimination 15. identificación Por lugar e of Percentage women who étnica have felt discriminated against on the basis of: 35 30 31.8 25 26.8 27.2 22.3 20 21.1 18.1 19.4 % 14.2 17.1 16.7 15.9 15 12.4 13.5 14.1 10 10.0 11.1 7.8 7.7 5 1.5 1.2 1.3 0 Skin color Language Economic Gender Clothing Age Sexual circumstances orientation Non-indigenous Indigenous Average Source: Authors’ calculation based on Perception Survey, 2014. 24 in local government (concejalas) has increased Figure Gráfico 7. Barriers 16. Tasa de to political participation participacion en la fuerza laboral significantly from 19 percent in 2004 to 43 per- Most common constraints faced by women (Question asked Por lugar e identificación étnica cent in 2010, according to the women’s rights open-endedly and up to two constraints were recorded) NGO La Coordinadora de la Mujer, in 2013 only Lack of time due to 39.2 7 percent of the country’s mayors were female. family duties 33.1 This suggests that one of the barriers to women’s Young children 19.5 19.4 political participation is prevailing norms about 19.4 Fear to be rejected the role of women in the political sphere. 16.0 Lack of political 17.0 knowledge 17.0 Time constraints are the main obstacle to Partner’s objection 16.2 17.9 enhancing women’s political participation. 9.5 Economic reasons 11.9 More than a third of women indicate that the 7.6 main reason they abstain from politics is their Political violence 9.3 family duties. This constraint, as well as the 0 10 20 30 40 fear of being rejected in the political sphere, is % more prevalent among non-indigenous women, No indígenas Indígenas while a larger share of indigenous women fears Source: Authors’ calculation based on Perception Survey, 2014. their partners’ disapproval and political violence (Figure 16). Only a small minority of Bolivian counterpart and 82 percent believe that women women (4.2 percent) believe that there are no represent female constituents better than male obstacles to women’s political participation, with politicians. This suggests that there is a clear an even smaller fraction of indigenous women demand and approval for female leadership in feeling this way. politics despite the many remaining barriers that female politicians face. Prevailing social gender norms further com- plicate women’s entry into and endurance In spite of high levels of representation at the in political life. Prevailing norms are reflected national level, elected or appointed female in women’s fear of being rejected if they partic- representatives still face discrimination and ipate politically (18 percent) and in the lack of a political violence. Quotas and gender parity partner’s support (17 percent).26 These concerns rules may be successful in getting women into are rooted in the traditional distribution of gender elected seats, but once in Parliament, women roles. Sixty-five percent of women think that polit- are pressured to vote along party lines, prevent- ically active women neglect their duties as wives ing them from challenging the status quo and and mothers and 21 percent are convinced that limiting their potential to promote women’s par- men make better political leaders than women. ticular interests and advance the gender agenda The share of indigenous women is slightly high- (World Bank, 2011a). There have been reports er for both of these indicators (Figure 17). At of harassment and incidents of public humili- the same time, 73 percent of women think that ation against female elected officials to force female politicians are as powerful as their male them to resign their offices so that the alternate candidate, usually a male, can take the post.27 26 There are striking urban-rural differences with regard In 2009, there were 120 known cases of such in- to these factors: in rural settings, 21 percent of women mention their partner’s objection as one of the main reasons for not being politically active, whereas in urban areas 27 Political violence appears to be more pronounced in the corresponding figure is only 14 percent. Fear of being urban areas. The Perception Survey finds that 9.2 percent rejected, however, is far more prevalent in urban areas of women residing in urban areas mention political violence where 21 percent of women have voiced this concern as a major obstacle to political participation. In rural settings, compared to only 13 percent in rural areas. the figure is 6.4 percent. 25 in 2009, they were able to develop autonomous Figure Gráfico 7. Social 17. Tasa denorms and political participacion en laactivity fuerza laboral Percentage of women who agree with the following forms of mobilization outside the indigenous Por lugar e identificación étnica statements: movement and create coalitions with the feminist movement and other NGOs, thereby gaining a “Politically active women neglect their 63.2 more prominent political role (Rousseau, 2011). duties as wives and mothers.” 68.4 The increased engagement and lobbying efforts “Female politicians are as powerful 72.2 contributed to the adoption of gender quotas as their male counterparts.” 75.8 for proportional and majoritarian elections. That “Men are better political leaders 19.7 than women.” 22.6 success demonstrated that, as Ionescu (2012) proposed, women’s collective interests may be “Female politicians better represent female 81.3 constituents than their male counterparts.” 82.2 better advanced by building coalitions among women’s groups and grassroots organizations. “Gender equality has been 69.6 achieved in politics.” 74.0 0 10 20 30 40 50 60 70 80 90 % Young and Married Women Non-indigenous Indigenous at a Particular Risk of Source: Authors’ calculation based on Perception Survey, 2014. Violence timidation, generally related to elective municipal Violence against women is one of the stark- offices (Htun and Ossa, 2010). The Association est manifestations of lack of agency and has of Women Councilors (ACOBOL) found that a far-reaching consequences for gender equal- total of 283 cases of political violence against ity and other development outcomes. Women women were reported between 2000 and 2011, who suffer violence cannot fulfill their potential. ranging from pressure to give up their positions Their ability to accumulate endowments (edu- and arbitrary withholding of salaries to physical, cation, health, and physical assets) and their psychological, and sexual acts of violence. After access to economic opportunities are restrict- a decade of discussion, the Bolivian government ed. They are more likely to have poorer health enacted in May 2012 a law prohibiting political and considerably lower earnings than women discrimination and violence against women. 28 of similar characteristics who are not subject to This law aims to establish prevention, attention, abuse, which in Latin American countries trans- and sanction mechanisms against political dis- lates into between 40 and 60 percent lower earn- crimination and is widely considered to be im- ings. Moreover, violence has profound impacts portant progress for women’s political rights. on economic and social welfare of future gener- ations. Hence, reducing violence against women Indigenous Bolivian women’s participation could go a long way toward improving gender in politics was strengthened as a result of equality and other development outcomes. campaigning to influence the content of the new Constitution. Like other indigenous wom- In Bolivia, gender-based violence remains en in Latin America, indigenous Bolivian wom- widespread, and young and married wom- en attempt within the indigenous movements to en are particularly vulnerable to all forms keep their unity and advance gender-differenti- of violence. According to the 2008 National ated interests. But during the political process Demographic and Health Survey (DHS),29 almost leading to the adoption of the new Constitution half of women who had ever been in a relation- 28 Ley contra el acoso y violencia política hacia las 29 Encuesta Nacional de Demografía y Salud, ENDSA, in mujeres, May 28, 2012. Spanish. 26 ship or married reported having suffered psy- chological, physical, or sexual violence at some point during the 12 months prior to the survey, either by their partner or by another person.30 Forty four percent of women report ever having experienced some kind of violence by their part- ner, and 8.5 percent report ever having experi- enced violence by a person other than their part- ner. Furthermore, 24 percent of all women who had ever been in a relationship or married stated that they had been victims of physical or sexual violence in the previous 12 months. Between 25 and 27 percent of women who had never attend- ed school or gone no higher than the secondary level had experienced physical or sexual vio- lence, while only 17 percent of women with high- er education had.31 The same tendency is not- ed in wealth quintiles, with violence being more common in the lower quintiles, and much less so in the highest. Younger women are at higher risk of abandonment, and restrictions on contact with than older women: 55 percent of women 20 to 24 family. During the first half of 2012, the police report having suffered violence in the previous registered 137 cases of violent death of women 12 months, compared to 41 percent of women and 61,251 cases of violence against women. All 45 to 49 years old. The Perception Survey sug- of this shows that gender-based violence is no gests that indigenous women face a substantial- isolated phenomenon. The Perception Survey ly higher risk of becoming victims of violence: sheds some light on the negative impact on over 60 percent reported such incidents, while women’s wellbeing: 41 percent of women victims the figure for non-indigenous women was 44 of violence report higher levels of fear and de- percent. Physical and sexual violence are just as pendency as a direct result of the violence they common in urban as in rural areas, but psycho- suffered.32 logical violence is more common in urban con- texts. Thirty five percent of all women had been Violence against women has been on the victim of psychological violence in the 12 months forefront of the political agenda in Bolivia in before the survey; 40 percent of women 20 to the last few years. The government declared 24 years old, and 30 percent of women between 2012 as the Year to Tackle All Forms of Violence 45 and 49 years old. The most common form of against Women (“Año de lucha contra todas las abuse mentioned was verbal insults and humilia- formas de violencia hacia las mujeres”). In March tion, followed by accusations of infidelity, threats 2013, the integrated law to guarantee women a life free of violence (No. 348) was passed after 30 The Perception Survey finds similar prevalence of violence: 49.7 percent of women have been victim of a process of consultations and discussions. The physical, psychological, or sexual violence at some point law aims to establish mechanisms, measures, in their lives. According to this data source, women in rural areas face a slightly higher risk: 51.3 percent compared to 48.8 percent in urban areas. 32 The effect is largest for women residing in rural areas. 31 The Perception Survey also highlights the importance The percentage of women reporting that they feel more of education in reducing the risk of becoming a victim of afraid and dependent is 44.8 percent, while in urban settings violence. only 38.3 percent of victims expressed such concerns. 27 set up in the Ministry of Justice to improve evi- dence-based decision-making. The legal framework designed to curb gen- der based violence will not have the expected results unless operational and service deliv- ery arrangements are in place to improve the efficiency and accessibility of justice institu- tions. . In addition, social norms on the accept- ability of violence continue to prevent survivors from reporting it (World Bank, 2011a). Survivors of gender-based violence are less likely to report and seek help if police officers, family courts, and health providers are not sensitive to GBV victimization and trauma. Geographical distance from courts can also work against resorting to institutions of justice. In Bolivia, the 2008 DHS and integrated policies of prevention, attention, data shows that women victims of violence are protection, and response for women who experi- more likely to seek help from friends or family ence violence. However, the state body that pro- (15 percent) than from any kind of institution (9 vides technical services for the implementation percent). If they do not seek help of any kind, and coordination of the autonomous governance the main reason is that they feel embarrassed or regime (Servicio Estatal de Autonomías) has humiliated and/or are afraid of retaliation. Among shown that implementation is limited. A costing the poorest quintile, the main reason that wom- exercise to estimate capacity among municipal- en did not report incidents was that they did not ities to provide the services for which they are know where to go. The Perception Survey con- legally responsible found that 77 municipalities firms that the majority of women (58 percent) by- (almost 23 percent) lack the financial resources pass institutional help in situations of violence. to provide even the minimum services required Indigenous women are less likely to turn to public by law. Moreover, 248 municipalities (74 per- institutions (19 percent compared to 22 percent cent) have the resources to cover only the ab- among non-indigenous women) and 61 percent solute minimum (SEA, 2013). Another important of them do not seek help of any kind (compared recent measure is the integrated law against the to 55 percent of non-indigenous women). trafficking of persons (No. 263),33 which resulted from strong demands from women’s organiza- Teenage Pregnancy tions based on the increasing number of young and adult women who go missing every year. Dropping, but Still Above Since its passage, the focus of the discussion Regional Averages has been the regulatory framework needed to Adolescent pregnancy is another dimension implement these laws. An information system on by which lack of agency manifests itself. violence (Sistema de Información de Violencia Teenage pregnancy is highly associated with Intrafamiliar) and a centralized reporting sys- poverty, lack of opportunities, and social exclu- tem for domestic violence in Bolivia are being sion. It has a direct effect on women’s economic opportunities by restricting school attendance, 33 Ley Integral contra la Trata y el Tráfico de Personas, No. 263, July 31, 2012. decreasing earning opportunities, and increas- 28 ing dependence on social protection programs, the country’s lowest rates of teen pregnancy and among other things. Adolescent girls who get childbearing, in contrast to non-indigenous teens pregnant and have a child are also at greater risk (Alfonso, 2008). for health complications as their physical condi- tions are not yet ready for this process (Azevedo Adolescent pregnancy is associated with et al., 2013). Delaying marriage and childbirth exposure to the risk of pregnancy, such as has been associated with an increase in the age at first sexual intercourse and early mar- accumulation of endowments and female labor riage/union. Becoming pregnant is not always a force participation (Miller, 2010). Under Bolivian conscious decision. But when it is, research has law, girls are allowed to wed at 14 and boys at found that teenagers take into account several 16, following a pattern in many countries that factors such as opportunity costs and their indi- girls can marry at a younger age. Younger age vidual and household characteristics when de- at marriage for women and a greater age gap ciding whether to become pregnant. Teenagers between spouses are associated with women’s who have high opportunity costs might be less lower bargaining power and higher risk of do- likely to engage in sexual activity and become mestic violence (World Bank, 2011a). pregnant. In Bolivia, studies have shown that ex- pected educational gains due to not bearing a Bolivia’s adolescent fertility rates have de- child as a teenager significantly affect the proba- creased in the last decade from 85 births bility of a teen becoming pregnant (Florez, 2005 per 1,000 women aged 15-19 in 2000 to 72 in and Alonso, 2008). 2012. However, these rates are still slightly high- er than the 2012 regional average of 68 (WDI, Knowledge and use of contraceptive meth- 2014). There are considerable differences be- ods among teenagers in Bolivia vary. The tween urban and rural areas. According to the DHS reports that 18 percent of all women aged Demographic and Health Survey of 2008, 36 per- 15-19 know of some contraceptive method, but cent of urban women between 25 and 49 years only 9.3 percent actually use some kind of con- old had a child before turning 20, while the cor- responding figure for rural areas was 47 percent. Educational attainment appears to be an even Figure Gráfico 7. Age 18. Tasaof de women at firsten participacion sexual relation la fuerza and laboral more important factor: among adolescents with first child birth Por lugar e identificación étnica Percentage of women ages 25-49 only primary education, 32 percent were already mothers or pregnant at the time of the survey. 46.7 Gave birth before Among young women with secondary schooling, Rural 23.5 the age of 2.6 the figure was 13 percent and only 4.3 percent for those with higher education. Figure 18 illustrates 35.6 Urban 17.2 the differences between rural and urban teens 1.6 in terms of the age of first sexual relations and 69.9 the age by which women had their first child. The Rural 49.1 sexual relation Age of first majority of births to teenage mothers in Bolivia 10.7 occurs outside of formal unions, placing them 64.0 and their children in vulnerable social positions. Urban 42.0 8.4 Contrary to the rest of Latin America, a study us- ing data from DHS 2003 found that indigenous 0 20 40 60 80 % groups in Bolivia are not more affected by teen 20 years 18 years 15 years sexual activity, pregnancy, and childbearing: Quechua and Aymara women appear to have Source: Authors’ calculation based on DHS 2008, Bolivia. 29 traceptive. For women 15-19 years who are in a relationship, 66 percent have knowledge of some contraceptive method, but only 41 percent use contraceptives. Rural teenagers have a low- er use of contraceptive methods compared to ur- ban teenagers, which could be explained by the greater access that urban dwellers have to sex- ual and reproductive information and services as well as contraceptives. While ethnicity does not play a role in the level of knowledge teenagers have about the fertility cycle, it does in the use of contraceptive methods: non-indigenous teenag- ers have higher rates of use of modern methods, while the majority of sexually active Quechua and Aymara teens have not used any type of family planning method, and when they do use one, it is usually a traditional one (Alfonso, 2008). 30 Summary of Findings and Further Research T his Note shows important progress in ticipation in the labor market, Bolivian women all three dimensions of gender equal- have a higher labor force participation rate than ity explored—endowments, economic the regional average. However, the female la- opportunities, and agency—and highlights bor force is largely concentrated in the informal some persistent challenges: and low-productivity sectors, which puts women at a disadvantage in terms of earnings. Also, •• The intersection of gender and ethnicity women in the labor market are still spending deepens the gaps in some development four times as much time as men on unpaid do- outcomes in Bolivia. mestic work. Bolivian women have made great •• Women feel discriminated against in dif- progress in terms of political participation at the ferent aspects of their lives, in particular higher levels of government, but are almost ab- indigenous women. sent as elected municipal leaders, while political •• Not having access to economic opportuni- violence and discrimination remains a serious ties limit women’s agency; while women problem. Finally, Bolivia has one of the region’s with higher levels of agency are able to highest levels of violence against women, but more fully take advantage of existing in recent years has developed a range of poli- opportunities. cies and laws aimed at protecting women from gender-based violence. In all of these domains, In education, access has improved for all women continue to experience discrimination groups, but educational attainment indicators and its consequences. The ethnic variable and show persistent disparities between men and the urban/rural divide are other important fac- women and particularly point to the significant tors of the Bolivian context that contribute to the negative impact of the intersection of gender complexity of these issues. and indigenous status. The maternal mortality rate is extremely high in Bolivia relative to other This final section is structured around the main countries in the region, and raises challenges findings related to the intersectionality between both in terms of access to health services for gender and ethnicity in general, providing an giving birth and the quality of those services. overview of the challenges Bolivian women face In spite of legal barriers for women’s full par- in each of the dimensions of gender equality cov- 31 ered in this Note. At the end of each description, gender and ethnicity play a dual role in indig- we present several areas with knowledge gaps enous women’s educational disadvantage. As that would require further research if there are discussed above, compared to non-indigenous to be sound recommendations for policy-making adult men, non-indigenous women and indig- and program design. enous men are between 9 and 10 percentage points less likely to complete primary school, Intersectionality but indigenous women are 27.3 percentage points less likely. Access to health care and The intersection of gender and ethnicity health insurance, as well as wage gaps, on the confers cumulative disadvantages on in- other hand, are areas in which indigenous sta- digenous women in Bolivia. However, the tus, rather than gender, is the key determinant discussion here also shows that the magnitude and where the size of the cumulative disad- of these disadvantages, as well as the relative vantage for ethnic minority women is small. In importance of gender and ethnicity in driving a addition, more indigenous women than non-in- particular disparity, vary by outcome. For ex- digenous women feel discriminated against ample, educational outcomes for indigenous on the basis of their gender. Nevertheless, it women are systematically lower than those of is important to note that even when there is indigenous men as well as compared to non-in- no gender- or ethnicity-based disparity in a digenous men and women, suggesting that given outcome minority groups may still be disadvantaged. Furthermore, being an indige- nous and rural woman aggravates vulnerabil- ities through the mechanism of lower endow- ments and economic opportunities. There are exceptional cases in which indigenous women demonstrate better outcomes than non-indige- nous women, such as in teen pregnancy and childbearing. In the political process leading up to drafting and adoption of the new Bolivian Constitution, and in the last two government administrations, indigenous women have claimed a larger role within the public sphere. Existing social and political processes appear to be enhancing voice and agency among eth- nic minority women. Going forward, it would be useful to explore potential explanations and underlying rea- sons for gender/ethnicity-based disparities, using additional qualitative and quantitative sources. This would provide a better under- standing of the intersectionality between gender and ethnicity, and could help identi- fy policies and actions to address existing disparities. 32 Education Women’s access to education in Bolivia has improved considerably in recent years, but challenges remain to making access uni- versal. Even though boy and girl rates of com- pletion of primary schooling and the female-to- male enrollment ratio for secondary schooling are close to parity, major gender gaps persist for indigenous and rural students. Indigenous and rural women have higher levels of illiteracy and school drop-out, which hinder their ability to take advantage of economic opportunities. Almost all Bolivian women (92 percent) would have wanted to study more. Since the level of education closely interacts with and in some cases determines life-span opportunities, de- nying women the opportunity to reach the level of desired schooling has detrimental repercus- health staff has increased over the last de- sions. The analysis presented in this Note has cades, but remain well below the regional av- shown that low levels of education are linked erage. The maternal mortality rate in Bolivia is to lower levels of agency, higher incidence of the second highest in Latin America and the unemployment, and greater difficulty in access- Caribbean, after Haiti. The limited progress ing the labor market. In terms of educational at- against maternal mortality may be partly due tainment, the intersectionality between gender, to a lack of quality in the institutions that de- ethnicity, age, and place of residence seems to liver medical care and services to expectant have a higher deterrent effect than the effect of mothers. Ethnicity and geographic location ag- gender alone. gravate the situation, with indigenous and rural populations’ maternal and child health indica- Further exploring the choice of education tors performing worse than those of non-in- by gender could yield rich new insights into digenous and urban populations. Both in rural women’s agency as well as labor force par- and urban areas, more non-indigenous women ticipation. Also, given persistent disparities give birth at health institutions than do indig- in the school completion rates of Quechua, enous women. The results of the Perception Aymara, and other indigenous women, future Survey suggest that discrimination faced in the research could focus on identifying the driv- health systems may play a role. ers of gender- and ethnicity-based schooling and dropout rates, especially in urban areas To strengthen the understanding of mater- where access to school infrastructure is rel- nal health issues, it would be important to atively better. explore the underlying reasons why rural women and indigenous women choose not Health to deliver their children in health clinics. Indicators that measure access to repro- Additional qualitative work might be required ductive and health services are mixed in to comprehend the subjective reasons be- Bolivia. The share of births attended by skilled hind these choices. 33 Economic Activity would study if they had more time at their dis- posal, showing that freeing up women’s time While female labor force participation is high could potentially contribute to household, and in in Bolivia, women are much more likely than the end country-level, productivity. The need for men to work in part-time or vulnerable em- flexibility, limited opportunities in the formal labor ployment, informality, and low-productivi- market, and the gender wage gap force women ty sectors. This is consistent with the findings to seek economic opportunities in the informal of both the Jobs (2013) and Gender Equality economy, at the cost of labor rights, pensions, and Development (2012) World Development and other benefits. Reports. They found that persistent gender seg- regation tends to trap women in low-productivity, It would be useful to collect and analyze ad- low-paying jobs, and that a range of factors keep ditional gender-differentiated information re- other women out of the labor market altogeth- garding time use, wage gaps, and paid versus er. Social norms, in combination with the level unpaid work to identify policies and actions of individual agency, can either help or hinder that could promote gender equality in the la- a women’s capacity to take advantage of ex- bor market. Also, further information on the isting opportunities, especially in the economic drivers of female concentration in small firms sphere. Gender roles and expectations reveal and on possible ethnic and intersectional themselves in the kind of sectors that employ disparities related to firm size could help in women and men, with women being over-repre- formulating these policies. sented in the Bolivian service sector and men in industry. In addition, women still carry the major- Agency ity of domestic and care responsibilities, which impose a serious burden on their time and ac- Gender norms and roles affect women’s op- cess to opportunities and resources. Almost 43 portunity to take full advantage of the nu- percent of women would work and 25.7 percent merous women’s rights guaranteed by the Bolivian legal system. Bolivia has made im- portant progress in providing legal guarantees such as gender parity norms for elections to con- gress, freedom from violence, equal pay for work of equal value, and protection from employment discrimination based on pregnancy, civil status, age, or physical traits. Still, domestic violence remains very high. Bolivian rural women tend to be more exposed to physical and psychological abuse, which in some cases is heightened by the lack of local institutions and means to sup- port and protect women. This is worrisome, as violence against women is one of the starkest manifestations of the lack of agency, and has far-reaching consequences for gender equal- ity and other development outcomes. A lack of agency manifests itself in low aspirations, and hinders upward mobility through lower invest- ments in human capital and production technol- ogy. Women victims of domestic violence can- 34 not fulfill their potential and are restricted in their economic growth. One area in which Bolivian women have done better is in expanded political representation in Parliament and in the Cabinet. Concerning adolescent pregnancy, another man- ifestation of the lack of agency considered in the Note, Bolivian women in general get pregnant at a fairly early age. Women in rural areas tend to have their first sexual experiences and first child at a very early age, while there do not seem to be any significant differences based on indigenous/ non-indigenous status. Further study of prevailing social gender norms about the role of women in the polit- ical sphere could help translate the national level progress in women’s political participa- tion into a larger role at sub-national levels. In terms of violence against women, norms concerning acceptance of violent behavior and the seeking of support are important, and it would be useful to more fully understand access to and use of existing services to pre- vent and respond to situations of violence. 35 Annex 1. Methodological Note and Results of the Census Data Analysis The scope of the census data analysis is limited because the data include only a few socioeco- nomic variables which can be analyzed as outcome indicators or included as control variables in regressions. In analyzing the link between basic outcomes and gender and indigenous status (both together and separately), the following model is adopted from Tas, Reimao, and Orlando (2013):34 Wi = f (α + β1 Gi + β2ETHi + β3 (Gi * ETHi) + γXi + ϵi) (Equation 1) where Wi is a well-being outcome for individual i (such as literacy, school completion, and access to health facilities and health insurance); Gi and ETHi are individual i’s gender and ethnicity, which takes the value of 1 for women and indigenous/Afro-descendants, respectively; Gi * ETHi is an interaction term depicting additional impact on indigenous/Afro-descendant women; and Xi is a vector of control vari- ables (including a dummy for the respondent’s age, age-squared, and a set of dummies for age cohorts). When the dependent and explanatory variables are all categorical—which is the case in this anal- ysis (see below)—the interpretation of the model becomes analogous to the difference-in-differ- ences and the outcomes of a “treatment group” can be compared against a “reference” group. Accordingly, the marginal effects on the three terms above capture outcome disparities for women, indigenous groups, and indigenous women with respect to non-indigenous men. Table 1 summarizes these links, whereas Table 2 shows marginal effects from probit estimation of Equation 1 with literacy and school completion rates as dependent variables. Measuring outcome disparities with respect to non-indigenous men Table 1. (reference group) When... Then... Shows outcome gap for… Gi = 1 Λ ETHi = 0 Marginal effect on Gi Non-indigenous women Gi = 0 Λ ETHi = 1 Marginal effect on ETHi Indigenous men Gi * ETHi = 1 ↔ Gi = 1 Λ ETHi = 1 Sum of marginal effects for Gi, ETHi and (Gi * ETHi) Indigenous women Fuente: Tas, Reimao y Orlando (2013) 34 See http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2373320 36 Determinants of literacy and school completion Table 2. (marginal effects, using non-indigenous men as reference group) Primary school Secondary school Literacy completion completion Women -0,027*** -0,091*** -0,060*** Indigenous -0,009*** -0,097*** -0,110*** Female* Indigenous -0,027*** -0,085*** -0,060*** Rural -0,052*** -0,229*** -0,223*** Age -0,002*** -0,010*** -0,009*** Obs. 7.129.134 4.354.518 4.354.518 Pseudo R. sq. 0,30 0,19 0,15 Source: 2012 Bolivia National Census of Population and Housing. Notes: Average marginal effects from probit regression are reported, with literacy and school completion rates as the dependent variable and non-indigenous/Afro-descendant males as the reference group. ***p<0.01, **p<0.05, *p<0.10. Regressions control for age-squared and age cohorts. Literacy rate for individuals 25 and older; primary and secondary school comple- tion rates for individuals 25 and older. Even though the 2012 census allows respondents to self-identify themselves as “indigenous or Afro-Bolivian,” it is possible to undertake a more detailed analysis of different indigenous groups based on self-identification into sub-categories. Extending the empirical framework present- ed above, the role of (self-identified) ethnicity and gender on well-being outcomes is estimated with the following reduced form model, which measures the penalty of being female and a member of an indige- nous group relative to non-indigenous males: Wi = f[α+β1Gi + β2Ai + β3Qi + β4Ii + β5 (Gi * Ai) + β6 (Gi * Qi) + β7 (Gi * Ii) + γXi + ϵi], (Equation 2) where Wi and Gi are as defined above; Ai indicates self-identification as Aymara; Qi is self-identification as Quechua; and Ii is self-identification as an indigenous person other than Aymara or Quechua, includ- ing Afro-Bolivians. The ethnicity variables are interpreted in the same way as summarized in Table 1, except that the disadvantages faced by different indigenous groups are captured separately. In addition, to account for the fact that specific indigenous groups reside in certain parts of Bolivia, the vector of control variables Xi includes 115 dummy variables indicating individual i’s province of birth, as well as age and age-squared. In the context of sub-group analysis of different indigenous groups, place of birth is a more refined control than rural/urban location, since it accounts for the possibility that location not only has an impact on education outcomes, but is also correlated with ethnicity. We know this to be true in Bolivia, since the Aymara are primarily concentrated in the department of La Paz where education outcomes are above the country average. Tables 3-5 show marginal effects from probit estimation of Equation 2 with literacy and school completion rates as dependent variables, estimated separately for different age cohorts to illustrate the fact that gender- and ethnicity-based education gaps are smallest among younger cohorts. 37 Determinants of literacy, by age cohort Table 3. (marginal effects, using non-indigenous men as reference group) 12-19 20-29 30-39 40-49 50-59 60-69 70+ Female -,002 *** -,001 *** -,008 *** -,022 *** -,061 *** -,110 *** -,184 *** Aymara ,000 -,001 -,003 *** -,008 *** -,028 *** -,057 *** -,161 *** Quechua -,002 *** -,003 *** -,005 *** -,017 *** -,053 *** -,108 *** -,205 *** Other indigenous -,007 *** -,011 *** -,020 *** -,037 *** -,066 *** -,115 *** -,190 *** Female* Aymara -,001 ** -,004 *** -,011 *** -,032 *** -,075 *** -,177 *** -,187 *** Female* Quechua -,001 *** -,010 *** -,026 *** -,040 *** -,051 *** -,083 *** -,061 *** Female* Other indigenous -,000 -,002 *** -,005 *** -,006 *** -,013 *** -,037 *** -,013 ** Obs. 1.431.126 1.531.898 1.170.985 853.078 622.001 419.181 350.089 Note: Regressions were carried out separately for each age cohort and also included controls for age, age-squared, and prov- ince of birth. ***p<0.01 **p<0.05 *p<0.10 Determinants of primary school completion, by age cohort Table 4. (marginal effects, using non-indigenous men as reference group) 14-19 20-29 30-39 40-49 50-59 60-69 70+ Female ,004 *** -,019 *** -,059 *** -,084 *** -,105 *** -,101 *** -,095 *** Aymara -,008 *** -,017 *** -,068 *** -,123 *** -,198 *** -,254 *** -,271 *** Quechua -,027 *** -,068 *** -,155 *** -,214 *** -,266 *** -,271 *** -,255 *** Other indigenous -,026 *** -,049 *** -,099 *** -,154 *** -,199 *** -,223 *** -,211 *** Female* Aymara -,035 *** -,081 *** -,158 *** -,213 *** -,192 *** -,142 *** -,105 *** Female* Quechua -,040 *** -,051 *** -,047 *** -,041 *** -,011 *** -,025 *** -,048 *** Female* Other indigenous -,011 *** -,011 *** -,019 *** -,022 *** -,005 *** ,005 ,010 * Obs. 1.045.863 1.523.423 1.170.985 849.742 622.004 419.185 350.090 Note: Primary school completion is defined as the completion of the first six years of schooling, following the current school system (primaria). Regressions were carried out separately for each age cohort, and also included controls for age, age- squared, and province of birth. ***p<0.01 **p<0.05 *p<0.10 Determinants of secondary school completion, by age cohort Table 5. (marginal effects, using non-indigenous men as reference group) 20-29 30-39 40-49 50-59 60-69 70+ Female ,006 *** -,023 *** -,039 *** -,061 *** -,068 *** -,061 *** Aymara -,044 *** -,122 *** -,173 *** -,225 *** -,232 *** -,183 *** Quechua -,134 *** -,195 *** -,228 *** -,248 *** -,228 *** -,179 *** Other indigenous -,100 *** -,120 *** -,158 *** -,180 *** -,183 *** -,144 *** Female* Aymara -,119 *** -,152 *** -,148 *** -,102 *** -,050 *** -,063 *** Female* Quechua -,056 *** -,033 *** -,009 *** ,029 *** ,019 *** -,027 *** Female* Other indigenous -,026 *** -,027 *** -,014 *** ,006 ,022 *** ,008 Obs. 1.523.423 1.170.987 849.742 622.004 419.185 350.090 Note: Secondary school completion is defined as the completion of the first twelve years of schooling, following the current school system (secundaria). Regressions were carried out separately for each age cohort, and also included controls for age, age-squared, and province of birth. ***p<0.01 **p<0.05 *p<0.10 38 Annex 2. Methodological Note on the Perception Data Analysis The Perception Survey on Women’s Exclusion and Discrimination explores the perception of indig- enous and non-indigenous women’s situation in Bolivia in the areas of education, health, and eco- nomic opportunities, focusing on discrimination and exclusion. The survey was conducted between December 2013 and February 2014 by the Bolivian NGO La Coordinadora de la Mujer. The universe of the survey is the economically active female population in Bolivia, i.e. women between the ages of 15 to 75. It is important to note that the survey contains two distinct samples. The first is the national sample with a total of 2,620 observations representative at the national, departmental, and urban/rural levels. This sample forms the basis for all analysis presented in this Note. The second consists of data collected on subsamples of the population, adding another 609 observa- tions to the dataset. The oversample of women of lowland and highland indigenous communities was included to capture the specific realities of these regions, but is not part of the analysis presented in this Note. The sample design considers each of the nine departments of the country as a sampling stratum. The municipality was treated as the Primary Sampling Unit (PSU). In total, the national sample covers 51 municipalities selected randomly controlling for the population weight within the department. Within each PSU, four clusters were identified and eight interviews were conducted in each cluster. The Perception Survey requires the use of weights to estimate the distribution of a given variable in the population. The weights are proportional to the population sizes and stored as variable peso1 in the dataset. Preceding the analysis, the weights are to be applied (using the “svyset” command in Stata) if one wishes to make inferences about the characteristics of the population. 39 References Alesina, A., R. Baqir, and W. Easterly. 1999. “Public Goods and Ethnic Divisions.” Quarterly Journal of Economics 114 (4): 1243–84. Alfonso, M. 2008. “Girls Just Want to Have Fun? Sexuality, Pregnancy, and Motherhood among Bolivian Teenagers.” Research Department Working Paper Series 615. Washington, DC: Inter-American Development Bank. Alkire, S., and J. Foster. 2007. “Counting and Multidimensional Poverty Measures.” OPHI Working Paper Series 7, Oxford Poverty and Human Development Initiative, Oxford, U.K. Alkire, S., and M.E. Santos. 2010. “Multidimensional Poverty Index.” Research Brief, Oxford Poverty and Human Development Initiative. Oxford, U.K.: University of Oxford. Andersen, L.E., and B. Muriel. 2002. “Cantidad versus Calidad en Educación: Implicaciones para Pobreza” Revista de Estudios Económicos y Sociales, No. 1, pp. 11–41. Aritomi, T., O. Origati, and M.B. Orlando. 2010. “Female Headed Households and Poverty in LAC: What are we measuring?” Paper presented at the 2008 Annual Meetings of the Population Association of America in New Orleans, April 17-19, 2008. Articulación Regional Feminista. 2012. “Informe Regional de Derechos Humanos y Justicia de Género.” La Paz. Ashwill, Maximillian, and Morten Blomqvist et al. 2011. “Gender Dynamics and Climate Change in Rural Bolivia.” Washington DC: The World Bank. AVERT. 2012. “HIV & AIDS Information from AVERT.org.” http://www.avert.org/hiv-aids-latin-america.htm#sthash. C29xiE2Q.dpuf Accessed on May 11, 2014. Azevedo, J.P. et al. 2013. “Teenage Pregnancy and Opportunities in Latin America and the Caribbean: On Early Child Bearing, Poverty and Economic Achievement.” Washington, DC: The World Bank. Baird, S., F.H.G. Ferreira, B. Özler, and M. Woolcock. 2013. “Relative Effectiveness of Conditional and Unconditional Cash Transfers for Schooling Outcomes in Developing Countries: A Systematic Review.” Campbell Systematic Reviews 2013:8. Beaman, L., E. Duflo, R. Pande, and P. Topalova. 2012. “Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India.” Science 335: 582–86. Beneria, L., and M. Floro. 2005. “Distribution, gender, and labor market informalization: A conceptual framework with a focus on homeworkers.” Rethinking Informalization: Poverty, Precarious jobs and Social Protection. Ithaca, New York: Cornell University Open Access Repository. Berthelon, M.E., and D.I. Kruger. 2011. “Risky behavior among youth: Incapacitation effects of school on adolescent motherhood and crime in Chile.” Journal of Public Economics 2011; 95: 41–53 40 Betron, M., and L. Fort. 2006. “Mainstreaming Gender in the Health Sector: Prevention of Gender-Based Violence and Male Involvement in Reproductive Health.” En Breve Series 2006 No 88. Washington, DC: The World Bank. Canelas, C., and S. Salazar. 2014. “Gender and Ethnicity in Bolivia, Ecuador, and Guatemala. Documents de Travail du Centre d’Economie de la Sorbonne. 2014.21. Paris: Sorbonne University. Camacho, S., N. Schwab, and R.P. Shaw. 2003. “Bolivia’s Reform to Improve Maternal and Child Mortality.” Washington, DC: The World Bank. CEPAL. 2012. “El Estado frente a la autonomía de las mujeres.” Santiago de Chile: CEPAL. Chioda, L. 2011. “Work and Family: Latin American and the Caribbean Women in Search of a New Balance.” Washington, DC: The World Bank. Coordinadora de la Mujer. 2013. Inputs provided for the development of the proposal “Understanding Agency by Measuring Women’s Perception on Exclusion and Discrimination.” La Paz. ———. 2012. “La construcción de un nuevo horizonte político: La despatriarcalización.” La Paz. ———. 2012. “Violencia Sexual en Bolivia: La dominación patriarcal sobre los cuerpos de las mujeres.” La Paz. ———. 2011. “Agenda de Equidad.” La Paz. ———. 2011. “Paso a paso. Así lo hicimos - Avances y desafíos en la participación política de las mujeres.” La Paz. ———. 2011. “Informe de Análisis de la Encuesta Nacional sobre Exclusión y Discriminación desde la percepción de las mujeres.” La Paz. ———. 2008. “Diagnóstico Nacional. Situación socioeconómica de las mujeres bolivianas y posibilidad de acceso a recursos financieros y económicos.” La Paz. ———. 2006. “La participación de las mujeres en la historia de Bolivia.” La Paz. Conexión Fondo de Emancipación. 2011. “Indicadores Nacionales 2010. Serie Bolivianas en Cifras 1.” La Paz. ———. 2010. “Linea base situacion de las mujeres en Bolivia por departamentos.” La Paz. Contreras, D., and M. Galvan. 2003. “Ha disminuido la discriminación salarial por género y etnia en Bolivia? Evidencia del periodo 1994-1999.” Departamento de Economía - Universidad Nacional de La Plata. Cortés, D., J. Gallego, and D. Maldonado. 2011. “On the Design of Education Conditional Cash Transfer Programs and Non Education Outcomes: The Case of Teenage Pregnancy.” CESIFO Working Paper 3531 (5): Economics of Education. Correa, N., and T. Roopnaraine. 2013. “Pueblos indígenas & programas de transferencias condicionadas: Estudio etnográfico sobre la implementación y los efectos socioculturales del Programa Juntos en seis comunidades andinas y amazónicas de Perú.” IFPRI, IADB, PUCP. Cunningham, W.V. 2001. “Breadwinner or Caregiver? How Household Role Affects Labor Choices in Mexico.” Policy Research Working Paper No. 2743. Washington, DC: The World Bank. Dani, A.A., and A. de Haan. 2008. “Inclusive States: Social Policy and Structural Inequalities.” Washington, DC: The World Bank. Das, M.B. 2013. “Exclusion and discrimination in the labor market.” Background paper prepared for the World Development Report 2013 on Jobs. Washington, DC: The World Bank. Das, M.B. 2012. “Stubborn Inequalities, Subtle Processes: Exclusion and Discrimination in the Labor Market.” Background paper for World Development Report 2013. Washington, DC: The World Bank. ECOSOC. 2010. Summary and recommendations of the report of the mission of the Permanent Forum on Indigenous Issues to the Plurinational State of Bolivia. Permanent Forum on Indigenous Issues, Ninth session, 19-30 April 2010. New York: United Nations. Farah, I., C. Sánchez, and F. Wanderley. 2009. “Políticas para la inserción laboral de mujeres y jóvenes en el Estado Plurinacional de Bolivia.” Santiago de Chile: CEPAL. Farah, I. 2002. “Familias Bolivianas y Trabajo de Hombres y Mujeres.” CIDES/UMSA. La Paz: Publicidad e Impresión Génesis. 41 ———. 2000. “Género y Pobreza: Una Relación Difícil. La Pobreza en Bolivia: Dimensión, Políticas y Resultados (1985 – 1999).” – Documentos de Trabajo. La Paz: Instituto Prisma. Fiske, S.T. 1998. “Stereotyping, Prejudice, and Discrimination.” The Handbook of Social Psychology, 4th ed., editors D.T. Gilbert, S.T. Fiske, and G. Lindzey. New York: McGraw-Hill. Flórez, C., and J. Núñez. 2001. “Teenage childbearing in Latin American countries.” Inter-American Development Bank Research Network Working Paper R-434. Washington, DC: Inter-American Development Bank. Flórez, C. 2005. “Factores socioeconómicos y contextuales que determinan la actividad reproductiva de las adolescentes en Colombia.” Revista Panamericana de Salud Pública 18(6): 388–402. Floro, M., A. Tornqvist, and E.O. Taş. 2010. “The Impact of the Economic Crisis on Women’s Economic Empowerment.” Working Paper 26. Washington, DC: American University. Galindo Reyes, F.G., J.M. De la Varga Salto, and A.M. Ciruela Lorenzo. 2012. “La mujer indígena campesina en Bolivia y su objetivo de igualdad. Propuesta de un mapa estratégico basado en valores cooperativos.” REVESCO 107: 98-128. Government of Bolivia. 2008. Plan Nacional de Igualdad de Oportunidades (PIO). La Paz: Government of Bolivia. Grepin, K. and J. Klugman. 2012. “Closing the deadly gap between what we know and what we do in women’s reproductive health.” Washington, DC: The World Bank. Hall, G.H., and H.A. Patrinos (editors). 2006. “Indigenous Peoples, Poverty, and Human Development in Latin America” New York: Palgrave Macmillan. ———. 2012. “Indigenous Peoples, Poverty, and Development.” Cambridge, U.K.: Cambridge University Press. Harbitz, M., and M.C. Tamargo. 2009. “The Significance of Legal Identity in Situations of Poverty and Social Exclusion.” Technical Note, 2009, 10. Washington, DC: Inter-American Development Bank. Hill, M.E., and M. Augoustinos. 2001. “Stereotype Change and Prejudice Reduction: Short and Long Term Evaluation of a Cross–Cultural Awareness Programme.” Journal of Community and Applied Social Psychology 11 (4):243–62. Hoff, K., and P. Pandey. 2006. “Discrimination, Social Identity, and Durable Inequalities.” American Economic Review. 96 (2): 206–11. Htun, M., and J.P. Ossa. 2010. “Political Representation in Hard Times: Indigenous Reservations and Gender Parity in Bolivia.” Social Science Research Center. Working Papers Series N40. Huarcacho, F.N. 2009. “Detrás del cristal con que se mira: Mujeres del Altiplano, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. Hunt, F. 2008. “Dropping out of school: A Cross-country review of literature.” CREATE. May 2008. Sussex: Centre for International Education, University of Sussex. Ionescu, L. 2012. “The Role of Women in Bureaucracies: Leadership, Democracy and Politics Economics.” Management and Financial Markets Volume 7(1): 138–143. Jackson, C. 1999. “Social Exclusion and Gender: Does One Size Fit All?” European Journal of Development Research 11 (1): 125–46. Latinobarometro. 2011. Data base Latinobarometro 2011. Santiago: Latinobarometro Corporation. Llanos, R. 2006. “Dimensión de Genero.” En Fretes-Cibils, V., M. Giugale, and C. Luff (editors). 2006. Por el bienestar de todos: Bolivia. Washington, DC: The World Bank. Marco Navarro, F. 2008. “El cuidado de la niñez en Bolivia y Ecuador: derecho de algunos, obligación de todas.” Serie Mujer y Desarrollo. Santiago de Chile: CEPAL. Martínez Franzoni, J., and K. Voorend. 2012. “Blacks, Whites, or Grays? Conditional Transfers and Gender Equality in Latin America.” Social Politics 19, 2012, 383-407. McGuire, J. W. 2013. “Conditional Cash Transfers in Bolivia: Origins, Impact, and Universality.” Work-in-progress paper prepared for the Comparative Politics Workshop, University of Pennsylvania, September 5, 2013. 42 Medeiros, M., R. Guerreiro Osório, and J. Costa. 2007. “Gender Inequalities in Allocating Time to Paid and Unpaid Work: Evidence from Bolivia.” The Levy Economics Institute Working Paper 495. Melgar, T. 2009. “Detrás del cristal con que se mira: Mujeres sirionó, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. ———. 2009. “Detrás del cristal con que se mira: Mujeres trinitarias, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. Miller, G. 2010. “Contraception as Development? New Evidence from Family Planning in Colombia.” The Economic Journal, 120 (545), 709-737. Molinas Vega, J.R., R. Paes de Barros, C.J. Saavedra, M. Giugale, L.J. Cord, C. Pessino, and A. Hasan. 2012. “Do Our Children Have a Chance? A Human Opportunity Report for Latin America and the Caribbean.” Washington, DC: The World Bank. Molyneux, M., and M. Thomson. 2011. “Cash Transfers, Gender Equity, and Women’s Empowerment in Peru,Ecuador and Bolivia.” Gender and Development 19: 207. Montaño Virreira, S., and C. Calderón Magaña (coordinators). 2010. “El cuidado en acción: entre el derecho y el trabajo.” Santiago de Chile: CEPAL. Morrison, A., and M.B. Orlando. 2006. “The Cost and Impacts of Gender-Based Violence in Developing Countries: Methodological Considerations and New Evidence.” Policy Research Working Paper 36151. Washington, DC: The World Bank. Movimiento de las Mujeres Presentes en la Historia, Bolivia. 2007. Series of documents from 2007 events, accesible on the webpage of the NGO La Coordinadora de la Mujer. La Paz: Asociacion La Coordinadora de la Mujer. Muñoz Boudet, A.M., P. Petesch, C. Turk, and A. Thumala. 2013. “On Norms and Agency: Conversations about Gender Equality with Women and Men in 20 Countries.” Washington, DC: The World Bank. Navarro, F.M. 2012. El Bono Juancito Pinto del Estado Plurinacional de Bolivia Programas de transferencias monetarias e infancia. Santiago de Chile: CEPAL. Ñopo, H., J.P. Atal, and N. Winder. 2010. “New Century, Old Disparities: Gender and Ethnic Wage Gaps in Latin America.” IZA Discussion Paper 5085. Bonn: Institute for the Study of Labor. Nostas Ardaya, M., and C.E. Sanabria Salmón. 2009. “Detrás del cristal con que se mira: Mujeres ayoreas- ayoredie, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. ———. 2009. “Detrás del cristal con que se mira: órdenes normativos e Interlegalidad. Mujeres Quechuas, Aymaras, Sirionó, Trinitarias, Chimane, Chiquitanas y Ayoreas.” La Paz: Coordinadora de la Mujer. Nussbaum, M.C. 2000. “Women and Human Development: The Capabilities Approach.” Cambridge: Cambridge University Press. PAHO. 2012. “Health in the Americas: Bolivia, Peru and Ecuador.” http://www.paho.org/saludenlasamericas/index. php?id=24&option=com_content&Itemid=&lang=en#ref34 Accessed on April 27, 2014. Plurinational State of Bolivia. 2009. Constitución Política del Estado. La Paz. Rousseau, S. 2011. “Indigenous and Feminist Movements at the Constituent Assembly in Bolivia: Locating the Representation of Indigenous Women.” Latin American Research Review Vol. 46 (2). Sakho, Y.S., T. Lunde, and M. Arribas-Banos. 2009. “Gender in Bolivian production: Reducing differences in formality and productivity of firm.” Washington, DC: The World Bank. Salazar de la Torre, C. 2007. “Género, etnia y clase. En busca de nuevas preguntas para la emancipación de las mujeres.” La Paz: Coordinadora de la Mujer. Salinas Mulder, S., M.D. Castro, and S. Zuazo Arana. 2010. “Análisis de los cambios en la equidad de género en Bolivia en el período 2004 – 2009.” La Paz: Coordinadora de la Mujer. Sanabria Salmón, Nostas Ardaya, and Coordinadora de la Mujer. 2009. Detrás del cristal con que se mira: Mujeres chiquitanas, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. 43 Schultz, T.P. 2004. “School subsidies for the poor: Evaluating the Mexican progresa poverty program.” Journal of Development Economics, 74 (1): 199–250. Sen, A. 1999. Development as Freedom. Oxford: Oxford University Press. Servicio Estatal de Autonomias. 2013. “Costeo de Servicios de la Ley No 348 ‘Ley Integral para Garantizar a las Mujeres una Vida Libre de Violencia.’” PowerPoint presentation prepared by the Unit for Analysis, Costing and Territorial Expenditures (Unidad de Analisis, Costeo y Gasto Territorial, part of the Direccion de Asuntos Autonomicos Economicos Financieros). La Paz, September 2013. Silva, E., and R. Batista. 2010. “Bolivian Maternal and Child Health Policies: Successes and Failures.” Policy Paper May 2010. Ottawa: FOCAL. SEDLAC. 2014. Data retrieved from the website with “Statistics by Gender.” La Plata: CEDLAS and the World Bank. Spears Brown, C., and R.S. Bigler. 2005. “Children’s Perceptions of Discrimination: A Developmental Model.” Child Development, 76(3), 533-553. Tas, E. 2014. “Gender, Indigenous/Afro-Descendant Status, and Outcome Disparities in Bolivia: Preliminary Findings from the 2012 Census.” Background document prepared as an input to this Note. United Nations Rule of Law. 2014. “Gender Equality.” Retrieved on May 19, 2014 at http://www.unrol.org/article. aspx?article_id=28. USAID. 2012. “Phase I Report on Gender in Bolivia.” Draft, January 2012. Villegas, H., and J. Nunez. 2005. “Discriminación étnica en Bolivia: examinando diferencias regionales y por nivel de calificación.” Estudios de Economía. Vol. 32 (2): 201-218. Wanderley, F. 1995. “Discriminación ocupacional y de ingresos por género.” Ministerio de Desarrollo Humano, Secretaria Nacional de Asuntos Étnicos, de Género y Generacionales, Subsecretaría de Asuntos de Género. La Paz: Government of Bolivia. Willman, A.M., and C. Corman. 2013. “Sexual and Gender-Based Violence: What is the World Bank Doing and What Have We Learned?” Washington, DC: The World Bank. Winkler, D. and S. Cueto (editors). 2004. “Etnicidad, Raza, Género y Educación en América Latina.” PREAL (Programa de Promoción de la Reforma Educativa en América Latina y el Caribe). Washington, DC: PREAL. World Bank. 2014a. “Voice and Agency.” Washington DC: The World Bank. ———. 2014b. “Gender at Work: A Companion to the World Development Report on Jobs.” Washington, DC: The World Bank. ———. 2013a. “Women, Business and the Law 2014.” Washington, DC: The World Bank. ———. 2013b. “Inclusion Matters: The Foundation for Shared Prosperity.” Washington, DC: The World Bank. ———. 2012. “The Effect of Women’s Economic Power in Latin America and the Caribbean.” Washington, DC: The World Bank. ———. 2011a. “World Development Report 2012: Gender Equality and Development.” Washington, DC: The World Bank. ———. 2011b. Peru Recurso Programmatic AAA—Phase IV: “Improving Health Outcomes by Strengthening Users’ Entitlements and Reinforcing Public Sector Management.” Washington, DC: The World Bank. ———. 2010a. Enterprise Surveys (http://www.enterprisesurveys.org). Washington, DC: The World Bank. ———. 2010b. “Women’s Economic Opportunities in the Formal Private Sector in Latin America and the Caribbean.” Washington, DC: The World Bank. ———. 2009. “Gender in Bolivian Production: Reducing Differences in Formality and Productivity of Firms.” Washington, DC: The World Bank. ———. 2008. “Whispers to Voices: Gender and Social Transformation in Bangladesh.” Bangladesh Development Series 22. Dhaka: The World Bank. 44 ———. 2005. “Gender Issues and Best Practices in Land Administration Projects: A Synthesis Report.” Prepared for PREM/ARD. Washington, DC: The World Bank. World Development Indicators. 2014a. Data retrieved from the World Bank’s World Development Indicators (WDI) Data Bank in May 2014. Washington, DC: The World Bank. World Economic Forum. 2013. “The Global Gender Gap Report 2013.” Geneva: World Economic Forum. Yáñez, E., R. Rojas, and D. Silva. 2011. “The Juancito Pinto conditional cash transfer program in Bolivia: Analyzing the impact on primary education.” FOCAL Policy Brief May 2011. Zabala Canedo, M.L., and Coordinadora de la Mujer. 2009. “Detrás del cristal con que se mira: Mujeres de Cochabamba, órdenes normativos e interlegalidad.” La Paz: Coordinadora de la Mujer. Zapata, D., D. Contreras, and D. Kruger. 2011. “Child Labor and Schooling in Bolivia: Who’s Falling Behind? The Roles of Domestic Work, Gender, and Ethnicity.” World Development Volume 39 (4): 588–599. Zoninsein, J. 2004. “The Economic Case for Combating Racial and Ethnic Exclusion in Latin America and Caribbean Countries.” Research Report. Washington, DC: The Inter-American Development Bank. 45