WHY DOES SHE MOVE? A ST U DY O F WO M EN’S M O BIL ITY IN L ATIN AMER IC AN C ITIES WHY DOES SHE MOVE? A ST U DY O F WO M EN’S M O BIL ITY IN L ATIN AMER IC AN C ITIES © 2020 International Bank for Reconstruction and Right and Permissions Development / The World Bank 1818 H Street, NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// Creativecommons.org/ This work is a product of the staff of the World Bank with licenses/by/3.0/igo. Under the Creative Commons Attribution external contributions. The findings, interpretations, and license, you are free to copy, distribute, transmit, and adapt this conclusions expressed in this work do not necessarily work, including for commercial purposes, under the following reflect the views of The World Bank, its Board if Executive conditions: Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the Attribution-Please cite the work as follows: Dominguez data included in this work. The boundaries, colors, Gonzalez, Karla, Ana Luiza Machado, Bianca Bianchi Alves, denominations, and other information shown on any Veronica Raffo, Sofia Guerrero, and Irene Portabales. map in this work do not imply any judgement on the 2020. Why does she move? A Study of Women’s Mobility in part of the World Bank concerning the legal status of Latin American Cities. Washington, DC: World Bank. any territory or the endorsement or acceptance of such Translations- If you create a translation of this work, please boundaries. add the following disclaimer along with the attributions: Nothing herein shall constitute or be considered to This translation was not created by the World Bank and be a limitation upon or waiver of the privileges and should not be considered an official World Bank Translation. Immunities of the World Bank, all of which are specifically The World Bank shall not be liable for any content or error reserved. 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 Photo credits: Bank. Views and opinions expressed in the adaptation Banco Mundial: p. 49 are the sole responsibility of the author or authors of the Bibiana Melzi: cover adaptation and are not endorsed by the World Bank. Sofia Guerrero: pp.6, 15 Third party content-The World Bank does not necessarily Shutterstock: p. 4 own each component of the content contained within the work. The World Bank therefore does not warrant that Istock: pp.24, 64 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 tore-use a component of the work, its 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. TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 GENDER AND MOBILITY: A REVIEW OF KEY ISSUES . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 3 METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.1 Research Design and Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2 Sites at a Glance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3.3 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4 FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 4.1 Constraints to Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.2 Constraints to Affordability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 4.3 Constraints to Acceptability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.4 Constraints to Accessibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Physical accessibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 A broader concept of accessibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 5 RECOMMENDATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.1 Recommendations to enhance transport acceptability . . . . . . . . . . . . . . . . . . . . 50 5.2 Recommendations to enhance transport availability . . . . . . . . . . . . . . . . . . . . . . 55 5.3 Recommendations to enhance transport affordability . . . . . . . . . . . . . . . . . . . . 59 5.4 Additional recommendations to enhance transport accessibility . . . . . . . . . . . . 61 6 CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 ANNEX 1: Research Tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 ANNEX 2: Description of field sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 1. Buenos Aires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 2. Lima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 3. Rio de Janeiro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 ANNEX 3: Extended Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Participant Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 ACKNOWLEDGMENTS 5 ACKNOWLEDGMENTS T his study was produced by Karla Domín- guez González (Gender Specialist), Ana Lu- iza Machado (Consultant), Bianca Bianchi Alves Latin America and providing cross-regional support throughout its implementation. The team gratefully acknowledges the Um- (Senior Urban Transport Specialist), Veronica brella Facility for Gender Equality (UFGE)1, which I. Raffo (Senior Infrastructure Specialist), Sofia funded the research, including the qualitative Guerrero Gamez (Senior Transport Engineer) data collection in Rio de Janeiro, Buenos Aires, and Irene Portabales (Urban Transport Spe- and Lima and the analysis for the report. We cialist). The team wants to thank the valuable especially thank Miriam Muller, UFGE Regional comments of the peer reviewers Arturo Ardila Window Manager, for her continuous support (Lead Transport Economist), Maria Beatriz Or- throughout the preparation of this research. lando (Lead Social Development Specialist), and Paula Andrea Rossiasco (Senior Social De- The research also benefited from valuable velopment Specialist). Research design and comments from Fatima Arroyo Arroyo (Urban implementation was coordinated by Ana Lui- Transport Specialist) and Leonardo Canon Ru- za Machado with the support of the excellent biano (Urban Transport Specialist) and the sup- team of local researchers involved in each of port of Amanda Tabanera (Consultant), Brittany the cities studied: Carolina Mazzi, Christiane Walters (Consultant), Lorena Sierra Valdivieso Severo, Frank Davies, Naila Takahashi, Pablo (Transport Analyst), Pedro Da Cunha Rego Lo- Nunes and Tuanni Borba in Rio de Janeiro, Bra- giodice (Consultant) and Licette M. Moncayo zil; Ana Iparraguirre, Mateo Nilesen Obieta, Pi- (Program Assistant). lar Llambías, Irina Philosoph, Tomás Bombau Our most profound appreciation to the key and Nicolas Giandini in Buenos Aires, Argenti- informants and to the women who shared their na; and Cinthya Astudillo, Angélica Casaretto, personal stories with us. The opinions, interpre- Giacomo Basilio, Johnnatan S. Saravia, Claudia tations, and conclusions expressed herein do Naganoma in Lima, Peru. The team also wants not necessarily reflect the views of the World to thank Shomik Mehndiratta for initiating this Bank, its Board of Executive Directors, or the research as Practice Manager for Transport in Governments they represent. 1 The multi-donor funded Umbrella Facility for Gender Equality (UFGE) supports initiatives that demonstrate a strong public good rationale and are catalytic for innovation and investment in frontier issues or areas with persistent gaps. In the World Bank’s Latin America and Caribbean (LAC) region, UFGE funding aims to contribute to advancing knowledge and gathering evidence related to women’s agency and economic opportunities. 1 INTRODUCTION 7 1. INTRODUCTION The ability to move freely is an essential human right and an enabler of individ- uals’ participation in social and economic life. In recent years, the transport sector has recognized its role as a catalyst for poverty reduction and has striven to promote inclusive transport for all. The United Nations has recognized the relevance of the transport sector for urban development and inclusiveness through the sustainable de- velopment goals (SDG11, Target 11.2): “By 2030, provide access to…sustainable trans- port systems for all, improving road safety, notably by expanding public transport, with special attention to the needs of those in vulnerable situations”. According to UN Habitat, two-thirds of the population will be living in cities by 2050 (UN Habitat, 2018). The efforts required to build sustainable and inclusive transport systems in the future will not be minor. A key concept orienting action towards the goal of transport inclusiveness is the concept of “accessibility.” Accessibility has been defined as “the ease with which an individual is able to access opportunities (e.g., employment, health care, education, etc.), within the city space” (Gomide, Leite, & Rebelo, 2005). The World Bank “It would take about two and a half hours has been focusing its accessibility efforts on the just to get to work, five hours on a bus to following dimensions: 1) planning transport to ride there and back every day. It wouldn’t increase access to jobs and health services, 2) be good at all. I quit my job because of the inclusive access, concerning transport access travel time. It’s inhuman. No way would I for the poor, for persons with disability and the continue to do that”. elderly; and 3) taking into account the specific 3 0 - 5 0 -Y E A R- OL D WO M AN, mobility needs of women. R I O D E J ANEIR O Efforts to improve transport inclusiveness will require paying attention to women’s ac- “I can’t say that I feel good when I am in the cess to and use of transport infrastructure and street waiting for a bus. But it is my duty. I services. Research from several countries has have to work…and if I don’t work, life gets consistently shown that men and women have more difficult. If I could choose, I’d stay at different mobility patterns and that transport home and not go to work. I think about that infrastructure tends to disfavor women. Wom- every day. I would stay at home taking care en rely on public transport and non-motorized of the children, taking them to school, to the modes of transportation (mainly walking) more doctor, everything…” than men. Women travel shorter distances and 1 8 - 2 9 -Y E A R- OL D WO M AN, within a more constricted geographical area, R I O D E J ANEIR O engage in more non-work travel, make more 8 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES multi-stop trips, run domestic errands, carry factors that discouraged them from traveling bags, and often travel accompanied by others and using the public transport system. A first (children, elders, etc.). Although these differ- group of women responded that they were ei- ences exist, decision-making processes related ther unable to afford fares or were deterred to the design and planning of transport systems by the quality of the transport services on of- systematically neglect them. fer, pointing to issues like lack of safety or over-crowdedness during rides. Interestingly, a In World Bank operations, measures to im- second group claimed that they had access to prove accessibility—including for women— public transport and could afford it but were re- have focused on improving what policymak- luctant to travel on it because they were afraid ers consider to be enabling characteristics of of traveling long distances alone—something transport infrastructure and services. For ex- they had not done before. While the first group ample, standard interventions to improve ac- of women faced external constraints to their cessibility involve expanding the coverage of mobility, the second group faced obstacles of an transport infrastructure, ensuring transport internal nature: social norms and negative mo- prices are affordable to the poor or integrat- bility experiences shaped negative self-effica- ing gender assessments into the planning and cy2 beliefs related to their mobility. This episode monitoring of transport operations. Measur- illustrates that factors other than transport ability is also essential. A standard accessibility availability and affordability have a significant indicator used in the planning phase to select role in shaping women’s mobility and work de- priority areas and evaluate the impact of a proj- cisions. If this is true, should those not also be ect is the number of jobs reachable within a 45- considered in transport planning to guarantee 60 minutes ride from project-affected areas. women’s accessibility in practice? But are standard efforts to improve trans- This study explores the range of constraints port inclusiveness (e.g., bringing transport to women’s mobility and access to economic infrastructure and services to areas with low opportunities in six low-income areas of ur- coverage) enough to ensure the take-up of ban Latin America through the lens of agen- transport services by poor urban women and cy3. We frame women’s agency in mobility as a thus improve their accessibility to jobs? Simi- woman’s ability to make and act upon decisions larly, is the standard accessibility indicator cap- related to her mobility and make full use of turing adequately the impact of a project on public transport systems. In this approach, we women’s accessibility, or only its potential? examine how factors shaped by gender, poverty, A short story from Mexico illustrates why and social norms interact with factors related it makes sense to ask the women themselves. to actual transport systems to either sustain A brief consultation conducted in a poor area or constrain women’s mobility and job acces- of Mexico City as part of a transport project sibility. The study relies on data collected from (Dominguez et al., 2016) asked women about 2 “Self-efficacy” is defined as the belief in one’s ability to influence events that affect one’s life, and to have control over how these events are experienced (Bandura, 1994).  3 “Agency” has been defined as the ability to make choices to achieve desired outcomes. This ability includes women’s control over family assets and numbers, freedom from domestic violence, freedom of physical mobility, and access to social support provided by community networks, families, and friends (World Bank, 2011). 9 1. INTRODUCTION 12 focus group discussions and 204 semi-struc- The study aims to contribute to the litera- tured interviews with women, men, and key in- ture on mobility in three main ways. First, we formants in three metropolitan regions: Rio de zoom into the diverse experiences of lower-in- Janeiro (Brazil), Buenos Aires (Argentina), and come women in urban Latin America, while Lima (Peru). most studies still focus on higher-income pop- ulations, or generalize women as a category. The study demonstrates that, apart from Second, by using an agency approach, we aim transport-related deficiencies, several factors to apply a more holistic version of the concept at the community, household, and individual of accessibility that goes beyond physical in- levels shape women’s capacity to make and frastructure and considers the full experience act upon decisions about their mobility. These of women’s mobility, from door to destination4. include fear of assault and sexual harassment Third, we use qualitative methodologies to help during the journey to work (between home and uncover the less tangible gendered processes bus stop, at bus stops, in vehicles); restrictive that shape women’s choices, while most stud- social norms related to women’s mobility and ies are quantitative. work; lower priority given to women’s employ- ment in households; limited financial resourc- Structure of the report. The study consists es; negative work and commuting experiences; of five sections in addition to this introduction. lower sense of self-reliance for traveling alone; Section 2: we discuss key findings from the lit- unequal division of care responsibilities; short- erature on women’s mobility and how they in- age of formal and informal care support; less form the study design; Section 3: we briefly pres- access to and use of vehicles owned by the ent the methodology; Section 4: we describe the family; and narrower social networks. These factors which shape women’s Agency in Mobil- elements represent a disproportionate burden ity and seek to show how these shape women’s for women in comparison to men in terms of decisions regarding mobility and work; Sections financial, social, and time costs. 5 and 6: we draw policy recommendations and conclusions. 4 Last mile connectivity is a concept used in transport planning to describe the movement of people and goods from a transport hub to a final destination (e.g. home). 2 GENDER AND MOBILITY: A REVIEW OF KEY ISSUES 11 2. GENDER AND MOBILITY: A REVIEW OF KEY ISSUES The literature on gender and mobility continues to expand, both in terms of meth- ods and locations. Most studies thus far are quantitative and have traditionally focused on identifying differences in travel patterns between men and women in multiple dimensions of travel behavior (e.g., mode of travel, distance, and time traveled) based on data disaggregated by gender5. A smaller but growing body of studies has employed qualitative or mixed methods to explore individuals’ real-life mobility experiences in a variety of contexts. While most studies target higher-income countries, research has also been emerging from a wider field (see Tanzarn 2008 on Kampala, Uganda; Srinivasan 2008 on Chennai, India and Chengdu, China; Elias, New- man, and Shiftan 2008 on Israel; Bhide 2016 on India, and Lecompte and Bocarejo 2017 on Bogotá, Colombia). Studies comparing gender differences in work activities (Vance and Iovanna, 2007; travel behavior have found patterns that are Duchene, 2011; CIVITAS, 2014); consistent across developed and developing • Women make more multi-stop trips, while countries: 5 men mainly follow more direct, linear pat- • Women travel shorter distances and are lim- terns from home to workplace (Murakami ited to a more restricted geographical area and Young, 1997; Root, 2000; McGuckin and (Rosenbloom, 2006; CIVITAS, 2014; Lecompte Nakamoto, 2005; Duchene, 2011; CIVITAS, & Bocarejo, 2017); 2014). • Women spend less time traveling and cover • Women are more likely to travel during off- shorter distances to go to work (Blumen and peak hours (Duchene 2011; CIVITAS, 2014); Kellerman 1990 for Haifa (Israel), Lee and • Women use cars less (Rosenbloom, 2006; McDonald 2003 for Seoul (Korea), Cristaldi Vance and Iovanna, 2007; Srinivasan, 2008; 2005 (urban areas in Italy), and Schwanen, Tanzarn, 2008; Polk, 2003;) and drive fewer Dijst, and Dieleman 2002 (Netherlands). miles than men (Rosenbloom, 2006). Women are more likely than men to work at home (Rosenbloom, 2006), and less likely to • Women use public transport more (Polk, engage in “extreme commuting” (Marion and 2003; Cristaldi, 2005; Rosenbloom, 2006; Horner, 2007). Women tend to locate their Vance and Iovanna, 2007; Srinivasan, 2008; businesses closer to home than men (Han- CIVITAS, 2014). son, 2003). • Women walk more than men (Srinivasan, • Women engage in more non-work travel 2008; Elias, Newmark, and Shiftan 2008 for while men’s mobility evolves around paid Israel; and Polk 2003 for Sweden). 5 Studies are mainly based on two categories of data: (1) large, national secondary data sets, with mobility data generally fo- cused on the journey to work, and (2) travel activity diaries which record out-of-home movements for all purposes. Travel diary data for a metropolitan area are accompanied in some cases by detailed spatial databases that describe the characteristics of the urban environment on a fine spatial scale (Hanson, 2010). 12 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES • Women cycle less than men (Pucher and backs weigh especially heavily on women in Buehler, 2008 for the United States, Den- rural areas (Lecompte & Bocarejo, 2017) mark, Germany, and the Netherlands; Cic- • There are fewer transport options for wom- locidade, 2016 for Sao Paulo, Brazil). en. Women rely more on circumferential • Women use cheaper and less efficient modes public transport routes, to which transport of transport (Duchene, 2011; Lecompte & Bo- operators give lower priority in comparison carejo, 2017; and Bhide, 2016). to the radial commuter routes that connect directly to downtown areas. A further issue Transport studies have identified different for women is that public transport networks factors driving these gender differences: often fail to provide adequate coverage of • Women’s lower financial capacity makes first and last-mile connectivity (i.e., between them more dependent on public transport transport hubs and home) (UN Habitat, than men (Peters, 2002). Greater participa- 2008). Privatized state-owned transport tion in part-time roles and lesser paid ca- companies might also lead to the exclusion reers, among other factors, weaken women’s of routes in peripheral areas, generally used financial capacity, and ultimately impacts by women low-income women, but consid- their ability to pay for public transport. ered to be less lucrative (Peters, 2002). • The unfavorable fare structure for multi- • Lack of personal safety on public transport stop journeys makes transport more ex- impacts women. Women tend to adjust their pensive for women. Women living in pe- travel patterns and behavior according to ripheral urban areas are poorly served by security considerations, including the fear public transport. Their greater participation of sexual harassment (see Box 1). These is- in care activities involves multi-stopping in sues can lead to curtailed mobility and the different places spread around the city and need for women to choose their travel times represent a costly outlay on fares (Noack, and routes carefully. The security concerns 2010). A study by Lecompte and Bocarejo are several: the type of transportation, the (2017) in Colombia suggests that women interaction with drivers and fellow pas- spend a higher percentage of their income sengers, the safety of the destination, and on public transport than men in their same whether the environment of waiting areas social groups, even if their trips are shorter is user-friendly or not (Loukaitou-Sideris & on average. Fink, 2009, based on quant data for the Unit- • Women face greater disadvantages in access ed States). Furthermore, the design of trans- to transportation. Since women walk more port-related spaces often fails to incorporate and depend more on public transport, poor safety features such as proper lighting and pedestrian pathways, and inconvenient ac- visibility that could mitigate the risk of as- cess to transport facilities impact more their saults on women (WB et al., 2015). daily journeys, which are already time-con- • Women spend more time doing household strained given their multiple responsibilities chores and on multi-stop trips to different at home and outside. Women also find it locations. Transport arrangements that are more difficult to access appropriate trans- unsuited to women’s needs are excessively port to informal job locations. These draw- 13 2. GENDER AND MOBILITY: A REVIEW OF KEY ISSUES BOX 1 Sexual harassment on public transport Sexual harassment on public transport is a global phenomenon. A recent poll in France reveals that 100 out of 600 women surveyed in two outer suburbs of Paris had experi- enced some form of sexual harassment on trains. A survey in Cairo and Alexandria in Egypt showed that 99.3 percent of women reported having experienced sexual harassment, most commonly in the form of touching or groping, in streets and on public transport - where women are most at risk (UN Women, 2013). Meanwhile, a 2012 survey in Mumbai (India) by the We the People Foundation showed that 80% of women had suffered sexual harassment on public transport. Common types of abuse on public transport range from leering, winking, and offensive gestures to unwanted touching and pressing against women and girls as well as indecent exposure and assault. Although sexual harassment in public places continues to be widespread, most countries lack formal mechanisms for reporting this kind of behavior on transport systems. A World Bank report on 100 countries shows that only seven countries possess appropriate legisla- tion to deal with sexual harassment in public (World Bank Group, IDB, ICRW. 2015). time-consuming, especially for those who tion of women in decision-making bodies in have to balance domestic responsibilities the mobility and transport sector 6 also con- with paid work (Noack, 2010). tributes to the issue (Duchene, 2011). • The development of most transport infra- Although there is increasing evidence on structure and services has historically been women’s mobility constraints, there are still gender-blind, i.e., does not consider the dif- some critical gaps in the literature. Most studies fering needs of women and men. The lack are from higher-income countries or generalize of data broken down by sex (WB, GWI, IDB, women as a category. Fewer studies focus on the & ICRW, 2015) and of appropriate meth- urban poor of developing countries, who face ods for collecting and classifying transport mobility contexts with specific characteristics data (Sánchez de Madariaga, 2013) are some such as greater social disparity, reduced avail- of the factors disabling gender-responsive ability of public transportation, and well-estab- transport systems. The under-representa- lished alternative systems to public transport 6 According to the Harvard Business Review, women make up 20% of engineering graduates, but nearly 40% of women with engineering degrees quit or never enter the profession. In one-on-one, in-depth interviews, the authors explored why and how women stayed in an industry where so many women drop out. While all agreed that engineering remains a challenging pro- fession for women, they managed to survive only providing they received appropriate support from colleagues, especially in the early stages of their careers. Fernando, Dulini; Cohen, Laurie; Duberley, Joanne. “What Managers Can Do to Keep Women in Engineering” Harvard Business Review, 2018. https://hbr.org/2018/06/what-managers-can-do-to-keep-women-in-engineering. 14 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES (e.g., taxis, bike-taxis, moto-taxis, etc.). (Uteng, mobility patterns.8 While this approach can 2011). The ways and the extent to which these provide insights into the root causes of mobil- characteristics affect the mobility of poor ur- ity patterns, it does not allow a more detailed ban women need more attention. For example, exploration of processes that generate women’s a recent study from Argentina describes the dif- decisions on their mobility. ficulties faced by domestic workers who travel Finally, there is a need to explore further how from low-income neighborhoods to more pros- mobility constraints affect women’s decisions perous areas in gated communities that are on other life outcomes more broadly. Women’s more easily accessible by private vehicles than pathways to empowerment are determined by public transport (Blanco and Apaolaza, 2018). different combinations of factors, one of them It will be essential to explore the differences in being access to economic opportunities. There travel patterns among women, depending on is a mutually reinforcing interrelationship be- their age, marital status, and income and mo- tween women’s agency, mobility, and access to bility contexts7. economic opportunities. Better economic op- In developing countries, studies have fo- portunities for women can promote women’s cused on diagnosing the travel behavior of agency by providing them with more money, women, and have not thoroughly examined and also by broadening and deepening their the gender processes that shape these be- networks and sources of information and sup- haviors. When conceptualizing gender and port. The enhanced physical mobility result- mobility, it is essential to explore not only fac- ing from having a paid job brings women into tors related to transport services and related contact with a new group of people at work infrastructure but also to see individuals in a and elsewhere (World Bank, 2012b). Converse- broader context. Locally specific gender norms ly, mobility constraints have a negative impact and beliefs inform the assignment of roles, sta- in terms of women’s career development and tus, power, and resources within families and income9. It is important to examine different communities. These elements can help to un- women’s mobility experiences so that policy derstand how and why gender can influence prescriptions can address the needs of all wom- mobility. As pointed out by Uteng (2011), survey en, while simultaneously countering economic data generally use variables as proxies for iden- and social disparities. tifying gender processes that shape women’s 7 Some studies have found no differences between women and men in terms of their travel activities (Gossen and Purvis 2005 for San Francisco; Vandermissen, Theriault and Villeneuve 2006 for Quebec). In the United States, studies have also failed to discover any gender differences in the length of work-related journeys among groups defined by race and ethnicity (McLafferty and Valerie Preston 1991 for the New York metropolitan area; Doyle and Taylor 2000). A recent study in Mumbai reveals that the proportion of women commuters whose mobility patterns resemble men’s is on the increase (Bhide, 2016). 8 For example, measures of ‘intra-household social relations’ are variables such as ‘marital status’ or ‘number of children at home. 9 An ILO report (2017) quantified the impact of women’s limited mobility on their ability to obtain and hold down jobs. It reckons limited access to transport and safety issues to be the main obstacles to women joining the labor market in developing coun- tries, and reducing their participation in the economy by 16.5%. 3 METHODOLOGY 10 10 The extended methodology can be found in the Annex. 16 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES In this study, we explore the connection between gender and mobility through the lens of women’s agency. Figure 1 depicts our conceptual framework. FIGURE 1 Conceptual framework of women’s agency in mobility Infrastructrure Social services Security Aspirations Self-e icacy Availability A ordability Acceptability Accessibility Accessibility to jobs (physical) Family and Experiences with peer support mobility and work Decision-making in the home Gender norms EXTERNAL INTERNAL Our starting point is the “Four A’s” frame- • Availability refers to connectivity and cover- work, which focuses on analyzing characteris- age of the urban public transport system. In tics of public transport systems and their rela- cities, public transport services are usually tionship to well-being and social inclusion. The unequally distributed, and lower-income ar- framework considers four attributes of urban eas are face low availability issues and low public transport systems (Gomide, Leite, & Re- quality of rail and bus services. belo, 2005; Pulido, Darido, Munoz-Raskin, and • Affordability refers to the cost of travel by an Moody, 2018): individual or family and the extent to which 17 3. METHODOLOGY people can afford to travel when and where actions, and constraints will also be shaped by they want. The financial cost includes the their level of income, their gender, and social direct cost of fares and the opportunity cost norms, among other factors. of potential consumption that is foregone in We frame women’s agency in mobility as exchange for mandatory trips. the ability to make and act upon decisions re- • Acceptability refers to the quality of urban lated to one’s mobility and make full use of transport infrastructure and user comfort, public transport systems. Our goal with com- safety, security, and reliability. bining these two frameworks is to explore both • Accessibility is the ease with which an indi- external and internal factors at the individual, vidual is able to access opportunities (e.g., household, and community levels that shape employment, health care, education, etc.), women’s decisions about transport use and within the city space, depending on avail- that, therefore, should be considered by policy- able transport infrastructure and services, makers in transport planning and implemen- personal characteristics, and time and other tation. Some of these constraints may be prac- constraints on their activities. tical ones such as the availability of childcare services, while others are more subjective such It is important to note that, for clarity pur- as social norms, socioeconomic aspirations, so- poses, the framework segments these four dif- cial network dynamics. Traditionally, transport ferent dimensions of transport attributes. In the studies have mostly focused on objective issues real world, these are interrelated and mutually that can undermine transport systems. causing. For example, the lack of transport cov- erage can determine high prices for the services that do exist. Transport prices can shape user 3.1. Research Design and behavior and influence the quality of transport Implementation modes. Also, availability, affordability, and ac- Study phases. We implemented the study in ceptability are critical for accessibility. several stages: (1) Study design and stakeholder engagement; (2) team training; (3) exploratory We then apply an agency approach to ana- data collection through focus group discussions lyze women’s decisions and set of constraints and participatory community mapping; (4) under each category of transport attributes. analysis of data from focus group discussions “Agency” as a broad term has been defined as and preparation of research tools for individual the ability to make choices to achieve desired interviews; (5) data collection through individ- outcomes (World Bank, 2011). This ability re- ual interviews with local men and women and quires an enabling circumstance on two levels. key informants; and (6) data analysis and final First, individuals need to be aware of their set write-up of results (see Annex 3 for details on of choices and have the internal disposition to each phase). act on one of them (the internal aspect of agen- cy). Second, the choices are only valid if he or Data collected. The study draws upon data she has the means to exercise action towards collected from 12 focus group discussions that choice and surpass exogenous constraints (FGDs) with women and 204 semi-structured along their path (the external aspect of agency) interviews with women, men and key infor- (Perova and Vakis, 2013). Individuals’ choices, mants (KIIs), in three Latin American metro- 18 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES ture (see Annex 1 for full interview guide). Key BOX 2 informant interviews with community leaders, transport agency representatives, transport Key to field sites providers, and academics provided contextual For practical reasons, the field sites will information. be referred to by abbreviations through- Criteria for selection of sites. Data were col- out the report: lected from six sites according to the following • Rio de Janeiro (RJ) dual criteria: • Buenos Aires (BA) 1. We selected two locations within each met- ropolitan region: one with “low accessibility” • Lima (LM) and one with “high accessibility”11 to explore • High accessibility area (HIGH) the variety of issues affecting individuals in different circumstances; • Low accessibility area (LOW) 2. All the sites had to be considered low-in- come areas within the given metropolitan politan regions (MRs): Rio de Janeiro in Brazil; region. Buenos Aires, Argentina; and Lima, Peru. We conducted two exploratory focus group discus- Criteria for selection of participants. In each sions in each research site (four per country), of the six sites, the sample was stratified by with women 18-29 years old and women 30-50 gender, age group, work status12, and distance years old. In FGDs, we used a participatory com- to work13. munity mapping methodology to probe discus- sion and produce a visual picture of women’s 3.2. Sites at a Glance14 mobility patterns and experiences (see Annex 3 for samples of maps produced) and identify Buenos Aires, Argentina relevant themes for individual interviews. Both focus group discussions and individual inter- Greater Buenos Aires is one of the world’s larg- views explored details of participants’ daily rou- est metropolitan areas and the third-largest tine and mobility; references to places outside urban agglomeration in Latin America. Great- their communities; experiences, attitudes, and er Buenos Aires has one of the most exten- decision-making processes related to mobility sive public transport networks in the world, and work; and aspirations and plans for the fu- including a suburban rail system of about 830 11 Accessibility is commonly measured as the number of job opportunities that can be accessed within 60 minutes for each origin-destination. In each metropolitan region, the team used a different tool to identify the level of accessibility of different research areas. See the extended methodology for additional information. 12 “Working” describes any individual who claimed to be engaged in a paid/income-generating (formal or informal) activity at least once a week. 13 “Working far” was defined as working outside of their neighborhood and surrounding areas. Rather than imposing a “hard definition based on kilometers, the team 14 We present a brief profile of research sites based on the available secondary data. Since data were not available at most micro level - i.e., for each high and low accessibility area, these are complemented with information collected from key informant interviews. 19 3. METHODOLOGY km with eight lines, a subway of about 52 km migrant” sector” in the city of Buenos Aires. with six lines operating within the city of Bue- Like the city’s other “villas,” the population of nos Aires, and a bus network operating about Villa 31 increased rapidly during the 1970s, 18,500 buses (CAF 2016). Half of its inhabitants with massive migration from the country’s use public transport daily, with the low-income interior to the cities. The latest estimates put population using public transport slightly more the population at 43,190. Half of the Barri- than higher-income groups. The most used os’s inhabitants originate from other coun- transport modes are buses (39.1 percent), rail tries, mostly Paraguay, Bolivia, and Peru. (6.3 percent), and subway (3.7 percent). For the • Ejército de Los Andes in Tres de Febrero, low low-income population (lowest household in- accessibility site: It is a community located come quintile), walking accounts for 36 percent to the northwest of the City of Buenos Aires, of all travel15. Lack of connectivity between the in the Metropolitan area. The neighborhood core and periphery poses socioeconomic chal- emerged as the result of the Villas Eradica- lenges leading to spatial segregation of poor tion Plan in 1968, aimed at relocating the in- and marginalized groups (World Bank, 2017). habitants of Villa 31 to the outskirts of the city. The first inhabitants settled in 1973. The area is known for its high levels of crime and violence. In both communities, the colectivo is the primary mode of transport. Informal modes of transportation include a combi in Barrio 31 that operates inside the community and “rem- ises truchos” in Ejercito de Los Andes, which are informal private vehicles that take passengers • Barrio 31 (previously Villa 31), high acces- (four at a time) from the neighborhood to Li- sibility site: Barrio 31 is known as the “im- niers, the closest transport hub. 15 Based on the ENMODO household mobility survey in metropolitan Buenos Aires for all travel stages performed on a typical weekday, 2009–10 (http://uecmovilidad.gob.ar/encuesta-de-movilidad-domiciliaria-2009-2010-movilidad-en-elarea-metro- politana-de-buenos-aires/). 20 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Lima, Peru Caja de Agua Chaclacayo High accessibility Lima Norte Low accessibility Low income Lima Centro Low income Lima Este Lima Sur Callao The Lima Metropolitan (Lima MR) region com- in the LMR. In 2015, about half of the 22.3 mil- prises the Metropolitan Municipality of Lima lion daily trips (including all modes and trips on and the Constitutional Province of Callao. It is foot) and the vast majority of public transport the residence of over nine million inhabitants trips in the LMR were made on the conventional (about a third of Peru’s population). The combi- system of largely unregulated and low-capaci- nation of its dispersed pattern of urbanization ty vehicles that compete for customers on the and the lack of an integrated network reinforces street (World Bank, 2015). social-spatial inequalities and limits access to In the Lima Como Vamos survey 2017, 60.9% urban centers where a majority of formal jobs, of respondents claimed that their daily com- education, health, and other services are avail- muting consisted mainly of walking, followed able. People in the two lowest income quintiles by the use of buses or combis. make 30% fewer trips for work, school, or other purposes than people in the top three income • Caja de Agua in San Juan de Lurigancho, quintiles, in large part due to high travel times high accessibility area: Caja de Agua is a and high travel expenses as a share of income16. community in the San Juan de Lurigancho district, one of the poorest districts in Lima. Lima MR still has a very limited mass tran- Around 90% of the population of San Juan sit network: one Bus Rapid Transit corridor (the de Lurigancho is in the very low-income cat- Metropolitano BRT) and an elevated rail line (as egory (50.1%) (the lowest income levels of known as Metro Line 1 or Tren Eléctrico), which the city). The area is known to have high lev- carry about 1 million trips per day but account els of crime and violence. In 2016, San Juan for only 9 percent of all public transport trips de Lurigancho ranked as second highest in 16 Source: JICA Origin-Destination Survey 2012 21 3. METHODOLOGY reported cases of “offenses against life and more confined neighborhood, since the dis- physical integrity” of all 50 districts in the trict is located in the relatively fertile valley metropolitan region17. of the Rimac, and as a result, is partly urban and partly rural. Although the socioeconom- • Chaclacayo, low accessibility: Chaclacayo is ic levels of the inhabitants vary, the study fo- a district located on the outskirts of Metro- cused on low-income people living in small politan Lima in the River Rímac valley, with communities scattered throughout the dis- a poverty rate of 10.1 % (seven points lower trict. The area is considered relatively safe. than Lima average)18. This study selected the In 2016, it ranked 41st of all 50 districts in the whole district of Chaclacayo rather than a metropolitan region. Rio de Janeiro, Brazil Jardim de Fonte Tavares Bastos Low accessibility High accessibility Low income Low income With a population of over 13 million inhab- er households in the RJMR make significantly itants, Rio de Janeiro MR is the third-largest fewer trips. Two explanations for the negative metropolitan region in South America. Like correlation between the number of trips and other RMs, Rio has a history of uneven urban income levels are the restricted access to mo- development marked by spatial segregation. torized transport and the high percentage of Public transportation in Rio is among the most income spent on transport: the poorest 10% of expensive in the world, with costs being partic- the population in the RJMR tie up around 22% ularly expensive for the poor. People from poor- of their incomes on transport compared to 17 https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib1534/libro.pdf 18 Source: “Mapa De Pobreza Provincial y Distrital 2013” de Lima. 22 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 15.5% of the 20% wealthiest (Scovino, 2008). A Janeiro. Queimados has a population of 138 2013 study showed that, for those living around thousand. Most of the households have a per the capital and the low-income population, the capita income of below the Brazilian min- region fared worse in terms of mobility indica- imum wage (R$ 1045 or approx. US$ 250) tors than any other region of Brazil (Pero and (IPEA, 2020) Queimados residents have a Mihessen 2013). mobility index that is 51% lower compared to the average index of the RJMR. Queima- • Tavares Bastos/Catete (high accessibility): dos’ travel time is the worst of the RJRM: 34% Tavares Bastos is a “favela” (slum) located higher than the metropolitan region average. in the southern portion of the city of Rio de The area has very high levels of crime and Janeiro, with a total of 1,100 residents. The violence. In 2016, it ranked first out of 309 latest gentrification process changed the Brazilian municipalities with over 100 thou- structure of the dwellings and the profile of sand inhabitants for the number of violent the population. Real estate price increases deaths19. encouraged the construction of buildings for rent as well as the “verticalization” of the Secondary data20 from the three metropol- favela (also due to restrictions placed on fur- itan regions confirm many of the gender pat- ther expansion of the area). The community terns documented in the literature: is situated on a steep hill, and commuters • Women travel less. In RJ, the average num- must take a moto-taxi, a minivan, or walk at ber of trips per day by women with family least 20 minutes down a cobblestone street incomes below three minimum wages is to reach the nearest transport hub. The 26% less than the number of trips taken by poor state of the road going up to the fave- men. In BA, women generate only 42% of the la, plus the lack of affordable transportation trips. Men in BA also tend to travel longer up and down the hill, is the main mobility distances than women. There are also sig- constraints that face residents. To climb and nificant differences in travel speeds: wom- descend from the community, residents can en travel at 8.62 km/hr, men at 10.93 km/hr. walk, use informal combis or moto-taxis. As for journey length, working women with When residents reach the bottom of the hill, children travel 4.77 km, while men travel there is ample public transport available in 6.72km. Speeds are 22% slower for women the Catete district. The area is generally con- than for men (9.7 km/hour vs. 12.2 km/hr, re- sidered very safe due to its proximity to a spectively). In Lima, there are differences be- special police unit headquarters. tween men and women in terms of average • Jardim da Fonte/Queimados (low accessi- trip times. Women spend less time on travel bility): Queimados is a city located approx- than men and typically tend to take a lot of imately 50 km to the northwest of Rio de very short trips (under 30 minutes). 19 In 2016, Queimados had a rate of violent deaths of 134,9 per 100,000 inhabitants. The figure is three times higher than the average of all municipalities with over 100,000 inhabitants. According to the study, the concept of violent deaths involves the sum of aggressions, legal interventions and violent deaths with undetermined cause, taking as reference the municipality of residence of the victim. Source: 20 This includes data from the Origin-Destination survey conducted in 2012 in RJMR (SETRANS, 2016); and the Pesquisa de Mo- bilidade e Pobreza in RJMR (ITRANS, 2003), Lima como Vamos survey (2017), ENMODO 2009 and the 2009 Household Mobility Survey for the Buenos Aires Metropolitan Region. RJRM (ITRANS, 2003). 23 3. METHODOLOGY • Women engage in more non-work travel. In to low-income population groups in urban Peru, RJ, while 48% of men travel for work or job Brazil, and Argentina. Security concerns meant seeking, only 19% of women’s trips fit this that fieldwork was restricted to specific loca- category. Studying, health, and household tions21 and hours of the day. Women and men errands account for 38% of women’s trips living in areas with higher levels of crime may (ITRANS, 2003). In the group of women who face additional barriers to their daily mobilities work at home, 38% of the trips relate to care that we do not fully capture in this research. duties. In BA, according to a WB study (Peral- Underage individuals were not included in the ta, 2014), once a family decides to have chil- samples because parental consent would have dren, the travel burden is assumed by the been needed, thus complicating recruitment. woman in the family (daily trips increase Furthermore, given the focus on the relation- from 1.57 to 1.78, while men’s trips remain ship between the labor market and mobility, we constant at 1.73 with or without children). chose not to explore the circumstances of the elderly and retired population. • Women walk more and rely more heavily on informal and public transportation. In The qualitative methods employed in this RJ, women use municipal buses more than study allow the exploration of a range of issues men (53.9% vs. 46.1%), taxi, motorcycle taxi, pertinent to a research topic from the stand- on foot, and the metro (subway). Meanwhile, point of those who experience social phenome- women in BA also use public transport more na (including those that are more subjective or heavily than men (50% vs. 37%) and walk sensitive). However, qualitative methods do not more (16% vs. 10%). In LM, men use their seek to provide an understanding of the distri- own cars to go to work twice as much as bution of the issues identified in the population women. Also, in Lima, women walk 7% more under study so that they are generalizable. than men. Teams also encountered some challenges • Women are less likely to drive cars. In RJ, due to the study’s multi-country approach, as survey data from 2012 shows that 57.1% certain discrepancies remained in implementa- of women did not possess a driving license tion. The research team dealt with these during in comparison to 36.2% of men. In BA, 28% the analysis phase either by not recording find- of men use their private car to go to work, ings on topics that had not been addressed while only 8% of women do so. Data is simi- equally in all three countries or by identifying lar to that for Lima, where only 6% of wom- such results as “country-specific.” en use a private car to go to work. The study’s objective is to identify the range of constraints faced by women in their day-to- 3.3. Limitations day routines. The study did not aim to provide Some topics and situations were not explored diagnoses of each location or to compare the in-depth in this study. The sample was limited different sites. 21 The sites had to be classified by the United Nations Department of Safety and Security (UNDSS) as safe for the purpose of conducting field work. 24 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 4 FINDINGS 25 4. FINDINGS This section of the report presents findings related to factors that shape women’s mobility decisions and preferences. It explores the interaction between commonly analyzed transportation factors (4A’s) with other external factors. Findings are based on data from 12 focus group discussions and 204 individual interviews with women, men, and key informants in high and low accessibility areas of Rio de Janeiro, Lima, and Bue- nos Aires metropolitan regions. The section is divided into four parts following the 4A’s framework: (i) availability; (ii) affordability; (iii) acceptability; and (iv) accessibility. 4.1. Constraints to Availability status of buses provided women with an ad- ditional sense of security because of the pos- Table 1 summarizes transport availability in sibility of making complaints to the bus op- all the research sites. In this study, different erators. In the areas considered to be unsafe, transport alternatives shaped women’s mobil- women who can afford to take public buses ity preferences: prefer this option to walking. In RJ, women • Walking: Women informants reported that also mentioned the existence of long-dis- walking was their preferred mode for short tance buses (known as “intercity buses” or trips for four main reasons. First, in areas its express version called “onibus executi- with limited or non-existent transport cov- vo”) in low accessibility areas as options for erage within the communities, walking is commuting to work in metropolitan region the quickest alternative. Second, regardless hubs. Generally perceived as higher quality of time constraints, women short of money services than municipal buses, their prices to pay for transport can always walk. More- were mostly considered prohibitive. over, walking means that they do not have • Microbuses and minivans: This transport to depend on unreliable public transport mode includes informal private transport as schedules and can estimate the length of well as vehicles regulated and operated by time that their journeys will take. Finally, private firms or cooperatives. In high acces- in areas where they felt safe, walking could sibility areas, services were typically short also be seen as a social and leisure activity distance, involving trips from the commu- and also an opportunity for exercise. nity to transport hubs. This choice of trans- • Buses: Municipal buses were the most fre- port was generally the only option for get- quently-used motorized mode of public ting in and out of the community, with the transport by women for short and medium most viable alternatives being moto-taxis distances, connecting passengers to differ- or walking. In low accessibility areas, micro- ent points within cities. One reason for this buses and minivans were longer distance is the predominance of bus routes and bus services, customarily used to travel from the stops compared to other modes of trans- centers of “satellite” cities to the larger met- port22. Although in many locations informal ropolitan transport hubs. Young women in alternatives were available, the “official” RJ, for example, prefer minivans over trains 22 Bus frequency in low accessibility areas was still very low. 26 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES for long-distance commutes, despite long times when other vehicles are not operating. queues, higher fares and traffic delays. An Women also like colectivos because of their added benefit is that, unlike trains, minivans relative comfort (i.e., few other passengers, have pre-fixed seats for passengers, which unlike buses) and as a way to avoid trans- act as a disincentive to harassment. port-related stress. In RJ, ride-hailing was in- creasingly used by women in both low and • Rapid Transit (Metro/Subway): The metro is high accessibility areas, given the more com- only available in high accessibility areas in petitive fares (significantly cheaper than reg- all locations. In all the sites, coverage is lim- ular taxis). Informants also mentioned that ited to a few lines. Women interviewed use passengers were sometimes discriminated the metro for short to medium trips in the against by official taxis (which also preferred downtown areas as an alternative to taking to avoid hilly streets). buses or walking. Opinions varied on the usefulness of the subway, depending on peo- • Moto-taxis: Moto-taxis were a feature of RJ ple’s location: in RJ, for example, informants and LM, where both informal and officially claimed that they used it when they could regulated associations of licensed moto-tax- afford it. is operate. These were considered faster, more practical (they take passengers direct- • Trains: In RJ and BA, trains are high-speed, ly to destinations), reliable (available out of direct, and relatively cheap for those taking hours), and cheaper than colectivos. However, long-distance trips. They are a good alterna- despite their advantages (much preferred by tive to minivans and intercity or executive men), moto-taxis were generally avoided by buses. In BA, however, the paucity of useful women because they were considered un- train routes meant low usage by our infor- safe, especially in places where pirate taxis mants. Trains were the mode of choice for operated. people whose homes or workplaces were near to train stations. The availability of transport within the communities was problematic in all the sites, • Taxis: Taxi services come in different forms. including in high accessibility areas. Some In all the sites, formal and informal “single communities were located on the top of hills or passenger” street taxis operate, with varying in areas difficult to reach by foot. From those availability and use. Ordinary taxis are used locations, reaching areas that concentrate jobs occasionally by women when traveling with usually involved two trips: the first segment children or on shopping trips involving carry- transporting passengers out of the neighbor- ing heavy bags. While official taxis were not hood to the closest transport hub and the sec- generally used by informants owing to their ond segment connecting passengers to city higher price, colectivos (taxi-buses) seemed centers. Public transport either does not serve to be a standard mode of transport, espe- internal routes in the communities at all (first cially in BA and LM. In BA, they are known segment), is only available at limited times (e.g., as remises truchos, accommodating multiple only during daytime), or is excessively infre- passengers along pre-set routes with multi- quent. In some sites, buses and minivans that ple stops. They fill transport gaps by taking normally circulate within and to/from commu- passengers in and out of communities at TABLE 1 Transport availability in research sites 4. FINDINGS RJ HIGH RJ LOW LM HIGH LM LOW BA HIGH BA LOW Buses Yes, outside the community Yes, inside the community. Yes, outside the Yes, inside the Yes, outside the community. Yes, outside the community. down the hill. There is one local bus community. community. line that cuts through the neighborhood and takes passengers to the nearest transport hub. Microbuses Yes, there is a short- distance Yes, outside the community. Yes. Yes. There Yes, inside the community. No. and minivans formal minivan service (also There are long-distance are minivan There are informal called “combi”) to enter/ informal minivan services services and short distance minivans exit the community, taking to take passengers to Rio de the authorized (“combis”), which run passengers up and down the Janeiro from the local city microbus through the neighborhood hill. Alternatives on this route center. To use this service, known as “El and can be used to enter/exit are moto-taxis or walking. passengers have either a Chosicano” to the community and reach 30-minute walk or take a take passengers the nearest transport hub. municipal bus, a moto-taxi, or to Lima. bicycle to the pick-up point. Metro/ Yes, outside the community. No. Yes, outside the No. Yes, outside the community. No. subway community. Trains Yes, but outside the Yes, but outside the No. No. Yes, but outside the Yes, but outside the community. Not widely used. community. Widely used. community. Not widely used. community. Not widely used. Licensed taxis Yes. Yes. Yes. Yes. Yes. Yes. Informal taxis No. Yes, individual. Not widely Yes, colectivos. Yes, colectivos. Yes. Remis Truchos operate Yes, colectivos, inside the used. in the community. community. Cars work as “unofficial taxis” to take They have variable prices people from inside the (according to distance) and neighborhood to pick-up are unregulated. points. The colectivos use fixed stops and prices and are unregulated. Ride-hailing Yes. Yes. Yes. Yes. services Formal Moto- Yes, inside the community. Yes, inside the community. Yes Yes No. No. taxi Informal No. Yes, inside the community. Yes Yes No. No. moto-taxi 27 28 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Given a lack of alternatives, women often have to walk long distances or take informal modes of transport to access public transport, implying higher exposure to risk. Walking be- tween communities and local public transport hubs can take over 30 minutes and involve hav- ing to go through unsafe and deserted places. “You get angry because you pay for a ticket Alternatives to walking were taking moto-tax- to the final stop of the route…I feel that is and other informal modes of transportation, drivers don’t respect us; they have no con- which women regarded as less safe. Addition- sideration for us because we are from the ally, research participants reported that even community…I once asked the driver, “why these supplementary forms of transport were don’t you go to where the bus is supposed to problematic, with drivers refusing to enter go [the stop]? I have paid for a ticket to the communities, or specific areas of communities, last stop…why don’t you take me there? He and often charging more to do so. just replied, “Get off the bus, slum dweller”. 3 0 - 5 0 YEA R- O LD WO MA N, BA H I G H Availability constraints impede women from accessing job opportunities. The service sector provided the most significant number “One day, we got back from work at 10 p.m. of opportunities for those interviewed, with at night, and the colectivo driver didn’t typical jobs among participants being cashier, want to come up here… he left us outside... waitress, store saleswoman, domestic worker, we had to come by Retiro, where we were among others. These jobs often entailed long mugged.” hours or night shifts. The lack of transport at 1 8 - 2 9 YEA R- O LD WO MA N, BA H I G H night meant women who work late or who are “extreme commuters” have even fewer alter- natives for getting home other than walking or spending part of their salary on informal trans- nities stop running in the late evening and early port. In contrast, men were more likely to sub- mornings. stitute motorized transport in late hours with walking. Discriminatory behavior towards poor women worsens transport availability issues. In RJ, ride-hailing services were increasing- In LM and BA, even when the official routes ly used by women. Although also problemat- served areas within or near to communities, ic, these services were becoming popular with bus drivers reportedly engaged in discriminato- women as drivers were more willing to enter ry practices, arbitrarily skipping stops or refus- their communities and other difficult locations. ing to pick up or drop passengers in these ar- eas. This issue was mostly reported by women, 4.2. Constraints to Affordability which could mean they are more often affected Women in research sites report spending a by such behavior. large proportion of their earnings on trans- port. Coverage and connectivity gaps in public 29 4. FINDINGS detriment of affordability and speed, especial- BOX 3 ly in areas with high levels of crime and vi- olence. For example, in Rio de Janeiro, men in Price discrimination for women in Lima’s moto-taxis low accessibility areas will walk or bike (cheap- er options) or use a moto-taxi (fastest option) to In LM HIGH, women reported difficulty reach the closest transport hub and then prefer in negotiating moto-taxi prices. Women the train (fastest option) to reach the area where living in the higher parts of the Lima foot- jobs cluster in central Rio de Janeiro. In contrast, hills risk being charged more for their the majority of women will take a bus for the trips. When they reach their destination first segment of the trip and a bus or a minivan and refuse to pay the high fares, women for the second segment. Women are more like- find that drivers often become rude and ly to perceive the cheaper and faster transport intimidating. To avoid such confronta- options commonly chosen by men as riskier in tions, many women prefer to walk, which terms of mugging, harassment, and accidents. makes their commutes more tiring and In this site, walking for longer than 5 minutes time-consuming. while carrying valuables is widely disregarded as an option by women. In LM and BA, some women reported episodes of sexual harassment transport mean spending significantly more on public transport and had, therefore, stopped time and money on trips. Women spend more of traveling by bus and metro altogether, choos- their earnings on getting around, face multiple ing to pay more for an informal colectivo. Many transfers, and have to depend more on informal women found the colectivo safer as they can transport. Women reckon that although the secure their own seats even if they traveled for cost of transport is reasonable, their key com- long distances. Other women deployed avoid- plaint is the accumulated high cost of transfers ance strategies that reduced the risks of harass- when they make multiple trips every day using ment, but which involved longer journey times different modes. Many women must combine (e.g., during rush hours, some prefer to wait for multiple modes to complete their journeys: less crowded vehicles to come along). walking, riding buses, and using a variety of informal vehicles such as minivans, moto-tax- The disproportionate amount of time and is, colectivos, or pirate taxis. These costs weigh money women spent on transporting them- heavily on their budgets. selves to work played a negative role in their accessibility to job opportunities. Many pay Prices of informal services can be negotia- for the costs incurred with complementary ble, creating disadvantages for women. In LM, and informal modes of transport from their women reported that although the price of mo- own pockets, which reduces their take-home to-taxi rides are negotiable with drivers, they pay and discourages many from working a long feel less able and comfortable than men to bar- way from home. Reduced fare schemes (such as gain and end up paying more (see Box 3). the Bilhete Unico) and employers willing to cov- When deciding which form of transport to er commuting costs, totally or partially, make it use, women (unlike men) tend to make trans- easier for women to work. port choices that prioritize their safety to the 30 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES BOX 4 Transport Affordability and Access to Opportunities in Brazil The cost of transportation can be a bar- “That’s why I have stopped studying…Trans- rier to accessing services, information, port is expensive... So what are we going to and employment opportunities. In Brazil, do? We have to work just to pay the fares… about a quarter of the cost of receiving these have gone up from 3 to 6 pesos, and health services in a hospital relates to they say they will rise again… to 11 pesos. transport expenditure. Data from 36 That’s really over the top”. countries show that low incomes are as- sociated with weak infrastructure-related - 3 0 - 5 0 YEA R- O LD WO MA N, BA H I G H mobility constraints (WDR, 2012). A study conducted in RJ as part of the urban “It’s dangerous [to cross the road] because transport plan concludes that people cars are speeding along and, since there are who earn less have to rely more on pub- no traffic lights, it makes it even worse to try lic transport, spend more traveling, and crossing... there is only a little space in the are less able to adjust their schedules middle of the road (...) very small and nar- (PDTU, 2015). The 2009 Household Bud- row, where you can wait (...) it’s even worse get Survey (POF) revealed that the 10% when you’re carrying babies, you feel like a poorest families in Brazil spent the most soldier, just standing there waiting to cross. significant proportion of their income on – WO MA N, 1 8 - 2 9 YEA RS , LM LOW transport (21.83%), while the richest 10% spent only 13.83% (IPEA, 2012). 4.3. Constraints to Acceptability many areas (e.g., areas perceived by women as very unsafe or too hilly23). In some loca- Participants, in general, evaluated the different tions, walking in and out of their community transport modes very poorly: with low levels involves climbing a maze of streets and stair- of comfort, reliability, and safety. Perceptions cases (challenging for those with young chil- vary according to the mode of transportation, dren, elderly or disabled people, and when and they are not limited to the service in itself carrying weight). Not only is this physically but expand to the surrounding communitarian tiring, but also and particularly problem- environment/infrastructure: atic in bad weather. Further impediments • Walking: Although the first choice by most include deficient pedestrian infrastructure women, walking is not a viable option in (lack of paved areas, unusable sidewalks24, 23 In this report, these were RJ HIGH, LM LOW and BA LOW. 24 In LM LOW, informants mentioned problems arising from irregular use of sidewalks by some local residents, who had built fences and set up businesses in pedestrian areas. Unauthorized parking also restricts women’s use of sidewalks. 31 4. FINDINGS crosswalks, and poor street lighting) and fear of accidents. • Buses: In some locations, buses were regard- ed as unsafe due to a higher risk of traffic ac- cidents, reckless driving, muggings, and the threat of sexual harassment onboard. These safety considerations applied to bus stops, “The train is a burden in our lives ... I’m ter- the ride itself, and the “last mile.” Other is- rified to leave here, to get a job downtown sues affecting women in all the sites were and be forced to take this train. It’s a big low bus frequency (i.e., longer waiting at bus sacrifice. You’re standing there with other stops), and poor lighting and limited security folks, stuck in the heat ... Very crowded, peo- at stopping areas, exposure to moving traf- ple on top of one another. Men take advan- fic on the streets, and difficulties of traveling tage of women. The train is lawless.” with small children or carrying heavy bags. - 3 0 - 5 0 Y E A R- OL D WOM A N , RJ LOW • Trains: In RJ, trains are considered uncom- fortable and unsafe (harassment and mug- gings) and are not preferred by most young bags or accompanied by children.25 women, despite them being the fastest way to reach central Rio de Janeiro. On the other • Informal taxis/Colectivos: Women feared for hand, while men perceived trains as unsafe their safety in the event of a driver’s miscon- and uncomfortable, such issues did not ap- duct due to the informality of the services. pear to deter them from using this transport However, many women considered colecti- option. vos a safe option in terms of sexual harass- ment because they were able to secure a • Microbuses and minivans: These transport fixed seat, and passenger rotation was limit- services were considered infrequent, over- ed along the ride. crowded, unsafe driving, and unreliable (holdups on departure). In Lima, microbuses • Rapid Transit (Metro/Subway): In RJ and BA, were seen as risky due to the high incidence women, on the whole, said they enjoyed of accidents and sexual harassment and fear the cleanliness and air conditioning, reli- of muggings when using this type of trans- ability, and low incidence of assaults com- port. In RJ, despite minivans being regard- pared to buses. However, the price was not ed as time-consuming (excessive road traf- affordable for everyone. In Rio de Janeiro, fic, long queues at access points, etc.), this the accessibility of stations was considered mode was preferred by many of the younger a major obstacle to using the metro, espe- women (i.e., they were able to secure a seat). cially for older women. Reaching platforms Another problem was that microbuses often often required long walks or the need to failed to stop for women carrying shopping climb stairs. In Lima, long queues and trav- 25 In some countries, fare policy can be the reason behind this discrimination. Some cities order cross-subsidies between regular users who pay full fare and children, students, elderly, even teachers that pay a reduced fare. The result is that drivers discrimi- nate against those that pay a reduced fare. 32 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES In all the sites, women spoke of the diffi- culties encountered by pregnant women or women with children or carrying shopping bags, etc. Transport gaps were critical in shap- ing women’s decisions to quit their jobs when they become pregnant. “I don’t trust moto-taxis 100%(...), they scare me (...) some of the drivers sometimes both- Crime and Violence against women er you and make obscene remarks. Within constraints to acceptability, safety and - 1 8 - 2 9 YEA R- O LD WO MEN, LM LOW security26 are major considerations women make when deciding which mode of trans- port to use (when alternatives are available). Except for two sites (RJ HIGH and LM LOW), the el times for relatively short trips discourage women interviewees reported being constant- the use of the “tren eléctrico,” especially in the ly exposed to situations of crime and violence mornings. The only existing metro line goes in their communities of varying severity: from from north to south, with limited access and pickpocketing, to assault and even murder. frequency (currently 360,000 passengers/ Women interviewed were more likely to con- day). For a journey that should take around sider the risks of accidents, criminal behavior, 8 minutes, informants report having to go harassment, and sexual assault when making by bus instead, involving a journey of up to travel choices. Their safety and security fears 2 hours due to heavy traffic. In all the loca- included the segments they had to walk to tions, however, assaults, gross overcrowding reach the stops, the period they waited at stops during peak hours, and sexual harassment as well as the entire ride in the vehicle. were mentioned as inhibiting the use of the Perceptions of security are significantly metro. different for men and women. For example, • Moto-taxis: Women in particular associated in RIO LOW, women reported a crime-ridden the use of moto-taxis with the fear of be- environment infested with gangs and exter- ing assaulted, robbed or sexually harassed mination squads (milicias), where “murders are (younger women), or out of concern for their becoming commonplace.” In contrast, men had physical safety (older women). In places more positive views of the community’s level of with regulated local moto-taxi drivers, with security–, some even describing it as a “peace- vehicles bearing an identity code, prospec- ful place.” Fear of crime and violence was not a tive passengers felt more confident to use significant issue recorded among male research this transport mode. participants. Additionally, in all the sites, both 26 According to the WB report, Cities in the Move (2002), security is defined as “vulnerability to intentional criminal or antisocial acts suffered by those engage in trip making”. This could include robberies or sexual harassment; whilst safety is more related to road safety. This report will consider the same definition of personal security. However, sexual harassment will be included as part of women safety, as this is the most frequent usage adopted by other international organizations which are a reference on gender issues such as UN Women. “Women’s safety refers to a range of strategies and policies which work to create safer environments for women and girls, often focusing on women’s insecurity and risk of violence in public spaces.” (Shaw, 2002). 33 4. FINDINGS BOX 5 A prevalent situation of violence in Latin America Latin America and the Caribbean is one of the world’s most violent regions. “ It is different for a man… my husband Insecurity stems from different factors, leaves at 5.30 a.m. to go to work… he just including drug trafficking, weak judicial goes off, and nothing happens to him. But and law enforcement lack of opportuni- as a woman, I am more afraid… more ties, among others. Research shows that insecure… you get it into your head that young people are more at risk to commit something is going to happen to you...it’s and suffer from this violence. Boys are different.” more likely than girls to be victims of 3 0 - 5 0 Y E A R- OL D WO M AN, B A LOW homicides, and evidence on perpetrators suggests that adolescence is the peak where people get involved in anti-social “The problem is at bus stops…When I’m behavior (Chioda, 2016). waiting somewhere for a bus, I am fright- ened because the stops are full of people, and thieves go by on motorbikes or bicycles BOX 6 to steal from you…So we try to stay near to other people because when you are alone, it Time of Day is scary. Frankly, I don´t even feel safe at the Women had specific preferences related bus station...” to the time of day when they undertook - 1 8 - 2 9 Y E A R- OL D WO M AN, B A LOW trips. Unlike men, women preferred traveling at off-peak hours when they “You can get killed for your purse. There are had the opportunity to do so, particu- no police around anymore… Previously you larly in mid-morning. This provided a could walk around and there was a cop on higher sense of safety due the sunlight every corner, but not now. The result is that I and fewer passengers (i.e. less chance feel insecure.” of harassment and robbery). Very early - 3 0 - 5 0 Y E A R- OL D WO M AN, B A HIG H or late hours were perceived as unsafe due to poor street lighting and having to walk across deserted areas, especially in low accessibility areas far from the city center. 34 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES sexes considered that women are more exposed to insecurity than men and are “easy targets” for crime and violence, especially while walking the “last mile” of their trips. Violence in the community impacts wom- en’s mobility in several ways: it restricts their “It’s disrespectful. We’re in there on our area of travel, increases their travel costs, con- way to work… many men think that just strains social interaction and work choices, because we’re women, we have to do what and increases the time they allocate to look they say, be abused by them. We women after their children. Women feel they can only are not there to enjoy an outing, we are travel at specific times or accompanied by oth- going to work, we are tired. If you let them, er people. In communities perceived as unsafe, the men will touch you and try on a lot of women are less mobile within their communi- things”. ties, and their trips are mainly work-related. 1 8 - 2 9 YEA R- O LD WO MA N, RJ LOW Sexual Harassment “There are crowds of people in the street, In all three cities, women reported experienc- and it can be uncomfortable because men ing or witnessing various episodes of harass- keep bothering you. One day my daugh- ment and assault in the streets and on public ter… was grabbed. (…) a woman called transport. Harassment and assault were an is- out to me: (…) “Didn’t you realize (…) the sue for women of all ages, including young girls girl has been grabbed, a man just grabbed (as reported by their mothers). On the street, in- her, and he started touching her?” (…) That formants said that the riskiest areas apart from made me so angry…so I shouted, “Leave the waiting stops were those near bars, often my daughter alone!” I then grabbed and full of men. As for public transport, the majori- punched him, saying (…) “I can’t allow you ty of the women were able to recount disagree- to do to my daughter what you have just able episodes on buses and trains. In RJ and LM, done”. women told of how these experiences impact- - 3 0 - 5 0 YEA R- O LD WO MA N, LM H I G H ed their travel behavior and costs in time and money. “I used to take the bus all over… once a Older women did not emphasize as much man kept pushing himself against me. I harassment as an issue that determined their threw my bag away, but the man kept trying mobility decisions. It is unclear whether this to kiss me. I told my friends to get off the bus related to less exposure or to their being more and walk…I was twelve.” accustomed to threatening situations. The fear - 1 8 - 2 9 YEA R- O LD WO MA N, RJ H I G H of sexual harassment was especially problem- atic for younger women interviewed. Women highlighted their feeling of help- lessness when bystanders did nothing to help. 35 4. FINDINGS In LM, women told of their attempts to defend themselves by shouting or even using physical force. In all the sites, women reported that it was common for bystanders to remain silent and not protect them against their aggressors, accusing them of being “crazy” or pretending. Although women recognized unwelcome “A friend who used to work with me... arrived sexual advances as an issue during their trips, in a desperate state at the nursery, because a frequent tendency was to regard men’s be- a guy was following her... She was desper- havior as normal or to blame the victims. In ate...she talked to him but he pretended not many cases, women expressed that it was their to hear. He didn’t stop following her and she responsibility to avoid triggering sexual harass- was forced to get off the bus before her final ment by wearing certain clothes and avoiding stop. It was bad…she was being followed, specific locations or hours. but nobody said a thing. Nobody. People Women’s fears are worsened by poor pub- prefer to turn a blind eye. Nobody wants to lic lighting and having to use informal trans- get involved in problems.” port, especially at night. Bus stops, often locat- - 1 8 - 2 9 Y E A R- OL D WOM A N , RJ HIG H ed in unprotected and poorly lit areas in the communities, are especially unsafe for women. Late at night, these places are deserted, and women become more vulnerable targets. Infor- mal taxis, buses, and moto-taxis were reported BOX 7 to be important supplementary forms of trans- Harassment in public transport port but were considered less safe than regular modes. Women informants commented that The study “Ella se Mueve Segura”, they could always complain to formal transport showed that 72% of women interviewed operators in the event of suffering problems felt unsafe using public transport (com- during their trips, but had little recourse in the pared to 58% of men) in Buenos Aires. In case of pirate operators. In RJ, the women in low the case of Lima, according to the 2017 accessibility areas who relied on moto-taxis survey of the organization Lima Como were more afraid and only accepted rides from Vamos, more than 30% of the women known drivers. On the other hand, women in interviewed had experienced sexual ha- a high accessibility area with a well-regulated rassment on public transport. However, moto-taxi cooperative used these services more while 2% reported this as being an issue, often and with less fear for their safety. only 10% saw this as the main problem The absence of an active police presence impeding access to transport services. in communities also contributes to the issue. In BA and RJ, women mentioned their distrust of police officers. In low accessibility areas, women claimed that the police were remiss in 36 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES their duties and chose to avoid problems. Nev- BOX 8 ertheless, in both sites, women associated the sharp rise in violent crime and violence to the The process of formalization of low levels police response, reckoning that po- vans in Rio de Janeiro lice presence had made the communities safer in the past. The informal transport system in Rio de Janeiro has seen a number of changes in the past decades. One of the main Women’s strategies to reduce modes of transport consists of minivans exposure to violence with a capacity of 10–15 passengers. Their operations started in the early The research found that familiarity with their 1990s by offering specific services to environments (places and people) enhanc- private groups. From the mid-1990s, es women’s mobility and sense of security. they expanded into the conventional Safety concerns affect both women and their transport market, competing with regu- family members and children. Women tended lar buses. The informal operations were to be more mobile and socially active in plac- gradually consolidated on ordinary bus es where they felt safe. This sense of security routes to downtown areas (Balassiano, arose from a greater police presence but also 1998). In 2005, the Rio de Janeiro Trans- from having good contact with neighbors and port Master Plan, designed by the State local business owners. In RJ HIGH, women’s Government, estimated that minivans fairly robust participation in community life, accounted for approximately 18% of together with strong ties with their neighbors, all the trips made in the metropolitan were perceived as protective factors. Women region on public transport. Many of described how they knew everyone in their ar- these services were regulated and - un- eas, were readily able to identify “outsiders,” like ordinary buses - operate in places and were alerted by neighbors in the event where access is difficult, such as narrow of any suspicious behavior. In BA, informants streets up in the Rio favelas (Sant’Anna mentioned that they were familiar with the et al., 2000). Since their expansion in the identities of local troublemakers and knew the 1990s, and despite State and Municipal unsafe areas in their neighborhoods, develop- government efforts to further regulate ing strategies to avoid certain places and the minivans, many pirate vehicles continue risk of having their homes burgled. to exist. Meanwhile, even authorized Developing friendly relations with bus minivans often fail to stick to their cor- drivers and street vendors was an important rect routes and pick-up/drop-off points. strategy to improve women’s commutes. In Other informal passenger vehicles also operate in Rio, such as ‘pirate’ buses and moto-taxis, but their passenger loads are minor compared to minivans (Balassiano and Alexandre, 2013). 37 4. FINDINGS RJ, women who knew the bus drivers were able to ask them to stop at convenient points along their routes even when there were no official bus-stops. This was a useful strategy in de- serted, possibly dangerous areas. As for travel by moto-taxi, women can contact drivers that they know personally by mobile phone to ar- “It’s dangerous at night, (…) we join other range rides at times which suit them best. In people, other women… “are you going to LM, as added protection, women made a point cross?” they ask… and “we have to cross all of chatting to other people waiting at bus-stops together.” (…) at the other end of the bridge, or with nearby street vendors. there are some bad people (…). There is a Women feel safer when traveling with woman with a (…) snack stand over there, friends or family members. Friends living in and I try to stay friends with her just in case the same area frequently organized joint com- one day I might need her to help me. This mutes or ride-sharing (for the few who private woman has warned me that things are always vehicles). In some places, women had regular happening there. You have to be careful.” arrangements with friends or family members - 3 0 - 5 0 Y E A R- OL D WOM A N, L M LOW to watch out for them when leaving or arriving at bus-stops. “When there is something weird going on… In RJ, while most women interviewed wel- because the [moto-taxi] drivers, they know comed women-only carriages (“vagões femini- us… so they say, “ah ma’am, ma’am...don’t nos”) as a positive measure, they complained go up by foot because there are some weird about their lax enforcement and the refusal of guys from another community walking up”, men to comply with the regulations. Younger and then, I say, ‘oh, I’m not going up walking.” women, particularly those in high accessibil- 1 8 - 2 9 Y E A R- OL D WOM A N , RJ HIG H ity areas, reported that they used the carriag- es regularly since they felt safer traveling in them. A recent study aiming at quantifying the “When you usually live a long time in the cost of sexual harassment in public transport same neighborhood, you already know the in Rio showcased that even if it has an impact hotspots, the [worst] times of the day, and reducing these incidences, commuters associ- then you can work around it.” ate women riding in the public space with more 1 8 - 3 0 Y E A R- OL D M A N B A HIG H openness to sexual advances, normalizing ha- rassment in public spaces (Kondylis et al., 2019). I know it [the pink carriage], but every time I got into one, there were plenty of men. {Laughs}. I didn’t see anything “feminino” there, just a sign with that word. It was just all pink, and it said “feminino” on top. - 30-50 YEAR-OLD WOMAN, RJ HIGH 38 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES BOX 9 “Vagões femininos” in Rio de Janeiro and new sexual harassment law in Brazil In 2006, Rio de Janeiro passed a law that introduced women-only carriages on trains and the metro, which became popularly known as “vagões femininos” (female carriages). This law, designed to solve the problem of sexual harassment on public transport during rush hours, has long been a controversial subject in public debate. Opponents of the measure argue that segregating women on public transport fails to address the root problem, which is the need to change social norms. Meanwhile, people who support the measure contend that, although not ideal, it is a worthwhile initiative designed to set aside safe spaces for women commuters to travel in peace. The consistent failure of male passengers to comply with this law remains an issue. Metro and train guards only enforce it on a few platforms in certain areas, and men often get into the pink carriages, especially at rush hour. In 2017, the law governing women-only carriag- es in Rio de Janeiro was regulated, with the Military Police made responsible for enforcing it. Violators are supposed to receive a warning on first offense and are liable to a fine if caught a second time. The fine in 2017 ranged from R$184,70 to R$1,152.77 (US$ 57-306) varying in the event of repeat offenses. We are unaware of any evidence that the new regu- lations are effective, or whether compliance issues have been resolved. Although several countries (e.g. Brazil, Japan, Egypt, India, Taiwan, Indonesia, Belarus, Philippines, Dubai, Korea and Mexico) have introduced women-only carriages, there is scarce evidence to date of their effectiveness in preventing women from being harassed. In 2018, a new law on sexual harassment (Law no. 13.718/18) was passed after the public outrage caused by the arrest and instant release of a man who ejaculated on a fellow bus passenger. The crime was considered a misdemeanor. The new law defines sexual harass- ment as a lewd act against someone without their consent that is committed by someone “to satisfy their own lust or that of a third person.” It carries a penalty of one to five years in prison. While skeptics show concern for adverse impacts the law may have on specific cases (for example cases of rape being typified as sexual harassment instead), the bill was widely received as a crucial step toward reducing impunity for cases harassment of wom- en in public spaces27. 27 https://www.huffpost.com/entry/brazil-sexual-harassment-law-public-transportation_n_5be48c33e- 4b0e84388957620?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAA- 7iOmRFwgHrS3oeccPKAL95osag4vG_pQgeKIn7fJduoZ2Lb6IMzeGZ8m-O677QjbdrEFzhB0ELlbmZQz9zVnHMGXdvC8l- lYGYgL_QIEine_Pe3Yl9uHS3fARw8klJdaB21koWaMWrBJkMhA7yV3U0xBn4KyXMRRE-H5mtLGP1x 39 4. FINDINGS 4.4. Constraints to Accessibility Physical accessibility Physical or spatial accessibility has a strong linkage with transport connectivity and land use. Land use factors can affect accessibility, What I dislike most is being far from the including density, mix, connectivity, and walk- center [of the city],.. downtown there are ability (Litman, 2017). Rapid transit, for in- job opportunities in universities, institutes, stance, can improve accessibility through phys- work centers. I know that the headquarters ical integration and a well-connected system of of large companies are down there, but if I different modes of transport, including feeder wanted to look for a job, I know that it’s not buses and non-motorized transport (Darido, feasible because of the distance from home. Moody, 2018). 1 8 - 2 9 Y E A R- OL D WOM A N , L M LOW The overlap between constraints related to gender and social class faced by women make access to jobs and opportunities more difficult, especially for women in low acces- sibility areas. Long distances, lack of available are more able to substitute segments of their transport services and infrastructure, and the trips with walking or biking, therefore saving shortage of local opportunities contribute to employers the costs of transport. the informants’ inability to benefit from job op- portunities and other services. A broader concept of accessibility In low accessibility areas, women report being discriminated against in job selection The concept of accessibility has been broad- processes for living too far away. Employers ening in the literature. From a traditional point reportedly prefer to hire people who live close of view, accessibility is defined as “the ease of to prospective jobs to avoid paying transport reaching desired destinations given a number costs and the risk of employees being always of available opportunities and intrinsic imped- late to work due to traffic problems in low ac- ance to the resources used to travel from the cessibility areas. origin to the destination” (Bocarejo & Oviedo, 2012). Under this definition, “impediments” re- Although both men and women are of- late to units of distance and time (Niemeier, ten turned down (by prospective employers) 1997). However, some authors, such as Geurs because of the distance of their homes from and Wee (2004) and Bocarejo and Oviedo (2012), suitable job locations, men are more adept at add to this definition other elements that can circumventing this issue. In RJ, women share influence a person’s ability to access econom- the view that, when competing for the same job, ic opportunities such as individual character- men are favored. Because they face less risk of istics, needs, skills, and information. Canon being mugged or harassed and have more time (2010) identifies dimensions of mobility-related available (due to the uneven care burden), men exclusion (physical, economic, temporal, spa- 40 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES tial, and psychological) as barriers for an indi- BOX 10 vidual’s accessibility. The relationship between social A deeper understanding of accessibility norms and access to economic from a gender perspective involves consid- opportunities ering not only the spatial structure of cities and transport networks but also other mobil- In general, Latin America has moved ity barriers related to how women experience towards closing gender gaps. For exam- transport. Gender-responsive accessibility anal- ple, the gender gap in primary educa- ysis must consider that women and men have tion enrollment has closed dramatically, different mobility patterns and daily activities, even having countries where girls are as well as the barriers imposed by differences achieving higher years of education in class, gender, age identities, and mental and when compared to boys (UNESCO, physical capacities (Levy, 2016). 2018). Moreover, Latin America has also Constraints to women’s mobility are not shown a relevant drop in fertility rates isolated issues but a combination of mutu- (Chioda, 2014). ally-reinforcing interrelated factors that limit However, these achievements have not women’s access to economic opportunities. necessarily translated into labor market In the previous sections, we have emphasized outcomes. Marriage status, education the set of constraints related to the features of level, and social gender norms are some transport systems. In this section, we explore of the elements shaping women’s deci- factors that are not strictly related to transport sions to access economic opportunities. systems (community, relational, and individual For example, traditional perceptions of factors) but which also affect women’s mobili- gender roles and responsibilities (men ty and accessibility. Since the study sought to as breadwinners and women as caregiv- clarify the relationship between agency in mobil- ers), can influence women’s decisions to ity and access to better jobs, our analysis also work and access specific jobs. Available throws light on factors that affect access to em- data from LAC shows that, even if family ployment opportunities. perceptions about gender roles have become more equitable (resembling Uneven division of care work: a highly OECD countries), they are still conserva- normalized constraint tive among the least educate (Chioda, 2014). The unequal care burden is seen as natural by both men and women. Housework and care activities are generally considered to be a wom- an’s lot, regardless of whether they work away from home or not. Men and women usually re- gard men’s domestic work as “help.” Women are reluctant to recognize this unequal approach as something to be negotiated and changed. 41 4. FINDINGS Transport issues exacerbate the time spent on care activities and lack of infrastructure. The shortage of transport within communi- ties and lack of school buses oblige women to walk several times a day to bus stops outside the community. Walks can take up to 30 min- utes and expose women to risk. Care options “My husband does very little housework, are limited or expensive. Although some older and I don’t ask him to help me because he women recognize that there are more daycare makes a mess…It’s better to avoid prob- options than in the past, most commented that lems and do it yourself.” these options were limited. - 1 8 - 2 9 Y E A R- OL D WOM A N , L M LOW Safety issues in these communities cause some women to extend the period of care for their children until adolescence. In BA, the “Two of them go to secondary school to- women report that although their children are gether, and two go to primary school, which old enough to take care of themselves, they be- is nearby. I have to go on foot because there lieve that a woman’s presence woman at home is no means of transport to get to school. ensures that the children do not fall prey to (...) it’s a bit isolated, and no colectivos go drugs or crime. there.” 3 0 - 5 0 Y E A R- OL D M A N , B A LOW An unequal care burden can affect wom- en’s health and restricts their ability to con- tinue their studies as well as hold down (and “That is when it all started... I left them at succeed in) their jobs. The added responsibility a time of their lives when they needed me of taking children to and from places entails most… and they learned about drugs… high costs in terms of time and money and The result is that the children are now leaves women with less time for leisure, edu- using drugs, although they don’t bother cation, and work. Although mothers may have anybody… but I cannot say that I‘m happy access to care facilities or other support, the with what they are getting up to. So, I stay hours are not always compatible with their jobs at home more now and keep better control outside the home. Working women admit that over them…I don’t go out much because the double burden of a paid post and child-rear- my house becomes a mess when I’m not ing impacts their health and their capacity to there… My husband goes out to work, and I improve their lives. In one interview, a woman stay home all the time.” complained about constant fatigue and being 3 0 - 5 0 Y E A R- OL D WOM A N , BA HIG H unable to do anything else, even to the point of having to give up her educational pursuits. In- deed, the household responsibilities of mothers are substantial, and it is no surprise that many admit to being “permanently tired.” 42 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Mutual childcare support networks en- hance women’s mobility. Such networks may consist of family members and neighbors (male and female). With access to a more extensive network, women are better able to work and undertake other activities. In RJ HIGH, it is com- mon for neighbors or neighbors’ older children “I’m tired all the time…Before I started to take turns taking smaller children to school working at the bakery, I was a “daily” and picking them up after school. In BA HIGH, domestic cleaner, working in five different the use of information technology has led to houses... [the unequal division of household the formation of informal support networks for chores] is no good for me because I used to women with limited other support. “WhatsApp study but had to stop. I want to go back to groups,” for example, are used to put mothers in classes, but now I can’t because I have no touch with others to arrange mutually-accept- time... I tried to finish school…go to col- able travel logistics for transporting children to lege, but I can’t…I used to fall asleep in the and from school, etc. classroom”. 3 0 - 5 0 YEA R- O LD WO MA N, RJ LOW Flexible work conditions can help wom- en to overcome the constraints of caring re- sponsibilities. In RJ, some women enjoy flexible We build a bond with neighbors ... I have working hours, payment by the hour, and some this neighbor who lives here.. she has two worked for employers who allowed them to take children and sometimes she has to go out their children to work. This was often the case somewhere, she asks me to keep an eye on of domestic servants (cleaners, etc.) working them for her. I say “Sure, of course.” We do close to home, who were allowed to stop work favors for one another, just like that” in the middle of the day, pick up their children 1 8 - 2 9 YEA R- O LD WO MA N, RJ H I G H from school and take them back to work while the mother finished her duties. “ I asked one of the mothers of the children on the bus who knew my daughter. I asked Partners’ negative attitudes-and her if she could keep an eye open because controlling behaviors-towards we were both alone. I would usually try to talk to the mothers of one of my daughter’s women’s mobility and work friends so that we could help each other out Most women interviewees assert that they if needed...” make work and mobility decisions freely and 1 8 - 2 9 YEA R- O LD WO MA N, BA H I G H with support from their partners. Partners’ support is, however, often conditional. Men expressed their support to women working out- side the home, acknowledging the benefits of an additional family income. However, men’s support is often contingent on women having 43 4. FINDINGS work schedules and locations that enable them to continue with other domestic responsibili- ties: cleaning, cooking, shopping, and childcare. There were many cases of women whose partners controlled their mobility and dis- couraged them from taking a paid outside job. Some men claimed they were concerned “If she leaves early for work she will not that women traveling “by themselves” would be able to fulfil her responsibilities, which be exposed to violence and harassment, or that means life gets difficult. Taking care of my children would be left unattended. Others were children is hard for me if she leaves early. openly jealous and suspicious that women who - 3 0 - 5 0 Y E A R- OL D M A N , L M LOW were out every day would be looking for anoth- er man to cheat on them. “It’s OK [for her to go to work], but not to The level of control partners exerted over work too long or hard like a man. She needs women’s mobility and daily activities varied, to stay home too. When you get home it’s with a few cases escalating to situations of good to have your wife or sister look after intimate partner violence (IPV)28 if the woman you. Women should also do the housework”. refused to change her work decisions. - 1 8 - 2 9 Y E A R- OL D M A N , B A LOW The support of families is essential for en- abling women to go to work. Support is pro- “I have always worked…Then I stopped vided in different ways, such as other family because my ex-husband wouldn’t let me members looking after their children, accom- work…He would say that women who had panying them to the bus-stop, helping with a job outside [home] were just looking for a commuting costs, or loaning money for them man. The truth is he was deceitful and didn’t to set up a small business. In RJ and LM, women want me to be independent” saw small-scale self-employment activities as – WOM A N , RJ LOW valuable income-generating activities. In RJ, for example, better-off families helped women to buy a market or street stall to enable them to “He didn’t like me to working but I worked work nearer home. anyway…We had constant fights…Once I put a knife on him. he came home…He wanted to hit me… I would not accept it, you know?...Oh, I don’t accept it, because we stay at home, we dedicate our lives to our children and husband, you work to also 28 IPV refers to any behavior within an intimate relation- provide to the house…to then get home and ship that causes physical, psychological or sexual harm to those in the relationship. This includes controlling get beaten…? behaviors, including isolating a person from family and – WOM A N , RJ LOW friends; monitoring their movements; insisting on knowing where she is at all times, and restricting access to financial resources, employment, education or medical care (WHO, 2012). 44 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES BOX 11 The impact of Gender Based Violence (GBV) on Women’s Agency “Oh, they give me lots of support … I must Different forms of gender-based vi- say that they speak highly of me, they value olence can impact directly women’s me a lot,…so much so that it leads to argu- agency. A study on Violence against ments between me and my husband. They Women in Latin America and the Ca- say: “My sister needs nobody to support ribbean (Bott et.al, 2012) showcased her; she’s an independent woman, she has that different forms of gender-based always worked, she always had her own violence are prevalent in the region. life, she bought her own house, she had her Evidence has also shown that violence own things. She doesn’t need to stand any against women has long-lasting conse- nonsense from anyone.” quences on women’s health, including 3 0 - 5 0 YEA R- O LD WO MA N, RJ H I G H physical injury, unwanted pregnancy, abortion, sexually transmitted diseases (Bott et.al.,2012) which have a direct impact on their self-esteem and deci- sion making. Freedom of violence is an Women’s paid work is given less essential domain of agency because of priority after marriage or children its intrinsic value as a human right and because its instrumentality to promote Men’s work is often given priority in house- gender equality (Klugman et.al., 2014). holds, while women’s is regarded as a second- This report recognizes that different ary income that is only needed if the family is forms of violence, including domestic short of money. Families are often obliged to violence, can constitute a barrier for pay someone to care for children while mothers women’s decisions to move and access go out to work. This option is rarely financially better economic opportunities; how- sustainable (since the outlay for childcare usu- ever, because of ethical considerations ally equals the mother’s earnings), and there is and a more general scope of the study, always the lingering fear that the children are this area was not explored in depth. As not receiving proper care by outsiders. can be seen in the report, sexual ha- rassment in public transport and urban Since care work is the woman’s responsi- spaces as well as controlling behaviors bility, and women’s salaries are often lower were identified as barriers for women’s than men’s, women are expected to give up mobility. Other forms of violence were work to devote themselves to care activities not explicitly manifested by the women whenever necessary. Alternatively, the cou- participating in the study. ple can decide that the woman will seek jobs 45 4. FINDINGS nearer to home with flexible hours (and lower pay). Low pay and childcare responsibilities are the main reasons for women giving up work, whereas men leave jobs when they are made redundant or to seek better pay. Women’s future goals more often prioritize looking after their families, whereas men as- “I want to [work], but only when my daugh- pire to better jobs and professional success. ter is older. My husband was thinking of This secondary nature some women attribute turning down jobs so that he could help to paid work after marriage is influenced by so- me. We ruled this out, because instead of cial norms that undervalue women’s paid work increasing our income we would be even and partners’ gradual enforcement of those poorer. norms through controlling behaviors and neg- - 1 8 - 2 9 Y E A R- OL D WOM A N , LM LOW ative attitudes towards women’s mobility and work. Most of the married women who were not working at the time of the study had start- “I’m willing to work, I love working, but ed working very early in their lives and expe- I cannot work because I have to look af- rienced a gradual detachment from paid work ter them. No matter how hard you try to activities after marriage or children. argue... men say “no, it’s we who have to work” and I have to go along with that.” To relieve care and transport burden, - 1 8 - 2 9 Y E A R- OL D WOM A N , B A LOW women prefer work proximity over quality en, and is reinforced by traditional gender roles. Negative commuting experiences influence Given that women generally earn less than women’s decisions to work closer to home, men (and their income is frequently seen as despite the possibly better job opportunities “secondary” to that of the male breadwinner) further afield. In all the study sites, women they prefer non-formal jobs with flexible hours argued that better jobs were available outside nearer home to accommodate care responsibil- their communities. Willingness to commute ities and bow to family pressures, and mainly meant that they would have access to formally to avoid having to commute to jobs with better registered jobs and better pay and conditions. professional prospects and benefits. In BA and In contrast, posts closer to home were regard- LM, women favored being self-employed, pos- ed as informal, unstable, low paid, and with no sibly setting up small businesses. These pref- fringe benefits such as paid holiday breaks, so- erences were very different from men’s: work cial assistance, and retirement pensions. for them was primarily a means towards higher Women prefer to work nearer home de- status and increased salaries regardless of lo- spite less attractive jobs and prospects. This cation. preference is shaped by the disproportionate Inadequate transport infrastructure plays burden that transport systems place on wom- a crucial role in families’ encouragement of 46 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Women’s social networks The range and strength of women’s social and support systems are critical determinants of their mobility patterns. In the low accessibility areas of RJ and BA, men and women can typi- cally rely on support networks in their neigh- “My husband asked me why I was think- borhoods. Some women, however, report that ing of working outside the home. He said, they generally distrust other people in the com- “Working outside means commuting. You munity. Others claim to have no friends and no- will have to take a train and put up with one to confide in, sometimes due to religious things like men stalking you. You aren’t beliefs. going to like it.” - 3 0 - 5 0 YEA R- O LD WO MA N, RJ LOW Personal networks facilitate access to jobs, especially for older and lower-skilled women. In LM and RJ, women’s networks are a valuable Does the car belong to both of you? Yes. (But way of obtaining information about job vacan- does only he drive?). Yes. I’m afraid to drive. I cies, especially for older women. Given the in- only drive a bit. formal nature of their work, personal recom- - 1 8 - 2 9 YEA R- O LD WO MA N, RJ H I G H mendations can count heavily towards securing a job placement. Women in paid work help family finances. In LM, businesses were often passed on within families, with women starting to work at a young age, following the example women’s limited mobility and opting out of of their mothers, selling food and clothes, or as paid work. An informant in RJ’s low accessibili- cleaners or babysitters. ty area said she dreaded commuting and, there- fore, hardly ever left her house, describing the Lower self-efficacy related to train journey to the city as an “unbearable ex- independent mobility perience.” Without a job, this woman said that she survived on a pension and financial income Exposure to risk, social constraints, fear, and support from relatives. Another informant negative experiences shape many women’s chose to work near home because her husband self-efficacy related to mobility. Self-efficacy did not want her to endure harassment on the refers to an individual’s belief in his or her ca- train. In the event, this man sold his car to buy pacity to execute behaviors necessary to pro- a small food stall to provide an income for his duce specific performance attainments (Ban- wife — both these initiatives involved having dura, 1994). Many women (although not the sufficient family resources, an option not avail- majority) claimed that they avoided traveling able to many. by themselves as much as possible. 47 4. FINDINGS The research also found that married wom- Bicycles are not a popular transport mode en were more afraid of walking alone and for women in all the sites, even in places using certain types of due to the stereotypes where men frequently use them. Older wom- of becoming a “too outgoing woman.” In RJ, en tend to ride bikes more than younger ones, community attitudes towards women’s use of as a cheap, healthy, time-saving option. Nega- moto-taxis were often negative (e.g., women re- tive aspects are bad weather, potholes, risk of garded as flirtatious). Married women, in par- accidents caused by car drivers, and the fear of ticular, were less likely to use moto-taxis. Social having their bicycles stolen. norms around the ideal behaviors for married women (e.g. to stay more at home), reinforced Social discrimination and lack of by their partners’ protective or controlling ac- tions played a role in shaping these women’s belonging perceptions of safety. Married women’s higher In some sites, women reported suffering dis- incomes (compared to single women) appeared crimination by people living outside their to steer them towards using less risky trans- communities. In BA, women expressed frus- port, gradually shaping their perceptions of tration, and occasionally resentment, that peo- safety. In RJ, many women confessed that when ple outside their community regarded them as they were younger and single, they would walk “delinquents” or “drug addicts.” Some women more uninhibitedly at night, even over long dis- indicated that their friends or relatives were tances. afraid to enter their neighborhood to partici- pate in social activities. This stigma extends to Less control over privately-owned their children, who find it challenging to play vehicles (cars, motorbikes, and with other children from outside the commu- nity. Many women (and men) said that they bicycles) had lost work opportunities after telling people where they lived. In RJ, a SEBRAE study (2013) Access to private transport is highly gendered. highlights that “people living in favelas have Motor vehicles such as cars and motorcycles readier access to the nearby dynamic jobs mar- are driven mainly by men. Decisions on the ket, but, in addition to living in areas with poor use of vehicles are almost exclusively down to infrastructure, often suffer discrimination due men. Many men use private transport as a re- to the social stigma attached to favela dwellers sult of frustration with public transport. They in general.” also find it more comfortable and time-saving. Notwithstanding the practical benefits, women, The discrimination experienced by women however, do not consider cars to be a viable op- in these communities has a direct impact on tion for them: they are less likely to own a vehi- their mobility by limiting their chances of se- cle or possess a driving license, and fear traffic curing employment and enjoying social net- accidents and ridicule. In BA and LM, a minority working opportunities. Some women find it of women drive motorbikes. challenging to apply for work outside the com- 48 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES munity unless they are willing to lie about their address. Although this issue also affects men, the experience of social discrimination com- bined with social norms that already encourage women’s limited mobility adds to a mindset in which women feel uncomfortable outside the community. This is well-illustrated by the ex- For us, everything beyond our neighbor- ample of one informant who said that she felt hood is another world… because we’re not like a pariah when visiting the center of Bue- integrated. I do not feel that I belong as a nos Aires, where people suspected her of being citizen… I carry the stigma around with a thief. me… It’s like living in a bubble inside your own neighborhood. 1 8 - 2 9 YEA R- O LD WO MA N, BA H I G H 5 RECOMMENDATIONS 50 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES The main challenge facing policymakers will be to design comprehensive interventions to address the multiple aspects of women’s mobility in low-income communities in de- veloping countries. Despite the many constraints outlined above, transport systems can play a facilitating role in women’s mobility. Changing these social/gender norms calls for medium-to-long-term interventions, but some short-term infrastructure-related measures could be taken. Collaboration between different stakeholders, besides transport-related, will also be needed to implement some of these recommendations successfully.29 In this section, we provide policy recommenda- • Improving data collection methods is es- tions to address some of the issues identified sential when planning transport and infra- in this study. These recommendations are not structure so that it responds to women’s exhaustive, and a holistic approach is needed needs. Gathering data disaggregated by sex when addressing women’s mobility and acces- and other factors intersecting with sex and sibility in a planning context. Several of these gender (such as income, family status, etc.) recommendations are based on promising improves transportation research and pol- practices since there is still not robust evidence icy. Traditional planning methods, such as about their impact. However, with proper con- origin-destination surveys, must incorporate text adaptation, they can be good entry points topics to capture women’s specific mobility to design interventions aiming at increasing patterns. Topics can include data on trav- women’s mobility and access to economic op- el speeds according to mode (even walking portunities integrally.29 alone or with other persons and carrying bags), trip chaining, and trip purposes.30 5.1. Recommendations to enhance Data collected through user satisfaction sur- transport acceptability veys should be broken down by sex, and the design of these instruments should reflect issues that affect women disproportionately, Gender-sensitive planning such as sexual harassment or transport de- • Boosting women’s participation in the trans- signs that fail to respond to their real needs. port sector provides for gender-sensitive plan- • Qualitative assessments of the mobility re- ning. Women’s involvement can be a con- quirements of poorer women in developing stant feature in decision-making on transport countries should form a crucial part of ef- planning, monitoring, and evaluation. forts to promote social and urban develop- 29 This report has shown that to increase women’s mobility and access to economic opportunities different angles need to be tackled (from infrastructure to social norms); this has to be taken into account when defining indicators related to increasing women’s usage of specific modes of transport and a robust gender analysis has to be done to support the definition of targets. 30 It is important for researchers and policy makers to understand that data broken down by sex does not aim to compare households headed by men or women, but to obtain information on different household members. The characteristics of female and male-headed households are not comparable, and relevant information on individuals within a household can be easily missed. While male-headed households generally include those in which women are married to men; female-headed households generally lack male adults. 51 5. RECOMMENDATIONS BOX 12 Enhancing women’s participation in transport sector decision-making According to the Harvard Business Review, 20% of engineering graduates are women, but almost 40% with engineering degrees either quit or fail to enter the profession. This re- sults in the transport sector remaining heavily male-dominated, which in turn means that women’s opinions as users of transport are not being heard. Transport operators have little incentive to respond to women’s needs. Women’s involvement in transport decision-making can be fostered by their increased participation in political institutions as well as employment in managerial positions in the transport sector. For example, in Rwanda and in South Africa, an increase in female lawmakers is related positively to the passing of progressive legislation that addresses women’s needs in different areas, including in the public transport area. Exposing girls to Science, Technology, Engineering and Mathematics (STEM) in schools and promoting paid internship programs with transport agencies can be transformative entry points. Furthermore, a study conducted in Malawi (Müller, Melibaeva, Machado & Casabonne, 2019) reveals the multiple and overlapping factors embedded in the social- ization, learning and hiring processes that result in low participation and high rates of attrition of women in the engineering and roads-related technical field at different stages of the career cycle, often called the “leaky pipeline” in policy discussions about women in STEM. The study calls for a career-cycle approach for realistically addressing the issue. ment. As well as detecting serious issues Prioritizing women’s safety in of transport infrastructure and services, qualitative assessments can reveal the con- planning and delivery text-specific set of socio-cultural issues that • Certain features of transport infrastructure impact women’s mobility. Also, it is possible have a direct impact on women’s sense of to assemble qualitative data on trips made exposure to risk. These include the availabil- and not made. Travel diaries, mobile phones, ity of transport alternatives and routes, low- or smart cards can also contribute to filling er prices, and a larger number of stops. data gaps and supply information on differ- ent mobility patterns, providing personal • Identify safe communal spaces, shops, etc., data protection laws are respected when us- for women to wait in near to bus stops con- ing this type of technology. sidered to be unsafe. Make sure they are well-lit and incentivize the set-up of small vendors. 52 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES BOX 13 Enhancing indicators used in the analysis of accessibility during planning Some tools have been used to conduct accessibility spatial analysis. The accessibility analysis calculates the number of jobs, educational centers, health centers, or other resources that are reachable within a certain travel time, either from a selected place or on average by all residents (Peralta, 2015). A common indicator used is the number of jobs reachable with 45-60 minutes time frame to illustrate the extent to which the urban transport system is serving a particular area or group of people (Darido, Moody, 2018). But is this accessibility indicator capturing well enough the constraints to women’s mobility? The indica- tor captures accessibility from the supply side, measuring, for example, the amount of people that could potentially use transport infrastructure if it is supplied in a given area. However, it does not capture infor- mation about where the assumption that people in that given area will become transport users actually translates into reality. A more accurate accessibility indicator could incorporate data from the demand side, related to mobility constraints faced by users in their decision-making processes. For example, the accessibility analysis normally does not provide enough granularity to understand pedestrian mobility within individual transport analysis zones and, therefore, does not take into account the frequent short trips which account for most of women’s mobility. An option would be to include data on pedestrian infrastructure and walkability. Moreover, jobs are normally defined as “formal job”, whereas low income women living in the peripheries are more likely to work in informal jobs. • Night buses could enable women to disem- • Implement non-discriminatory measures. bark wherever they find it most convenient. Concession agreements between the Regu- In Vigo (Spain), women can ask bus drivers lating Agency and ride-sharing companies to stop at any place where they feel safe. could include rules to forbid drivers from excluding certain areas (by providing mon- • Appropriate technology could also be ap- etary incentives if necessary) and install plied to ensure compliance with rules, reg- panic buttons in vehicles to increase opera- ulations, and controls, as well as enhancing tor and user safety. operational management systems to ensure that bus routes are respected. For other • Consider introducing controls and incen- forms of transport such as taxis, moto-taxis, tives to (as part of the ride-sharing conces- authorized minivans, etc., to consider simi- sion agreements) to make drivers deliver lar anti-discriminatory measures. passengers to the peripheral areas of cities. • Examine the possibility of running flexible services such as “demand-driven” transport. 53 5. RECOMMENDATIONS Using technology to enhance women’s BOX 14 safety Participatory tools to improve • Mechanisms to facilitate the reporting of environmental design for sexual harassment cases could include help- preventing violence lines or mobile applications for reporting to the appropriate response services offenses Safety audits depend on data gathering committed on public transport. Dissemina- from users to provide authorities with tion of vehicle number-plates could also be recommendations for ways to focus made mandatory by the regulatory agencies. on danger hotspots, improve women’s In RJ, women reported feeling safer travel- security and boost their sense of owner- ing on regulated transport because they had ship of public space (UN Habitat, 2008). the company’s number to call in the event Safety audits can be carried out in of abuse. The “Bajale al Acoso” initiative by different ways, including for example the Quito Integrated Transport System is a safe walking, the use of security-re- text message mechanism that generates a lated technology, etc. One approach response protocol by the Operations Com- is Safetipin, a mobile application de- mand Center when activated. veloped in India that allows users to • Such technological platforms need to go collect data and assess their perception hand in hand with capacity-building for the of safety by gathering information on staff of the relevant “response” authorities, parameters such as lighting, open- which are perceived by survivors of abuse ness, visibility, crowds, security, route on public transport as lacking credibility. In pavements, footpaths, availability of Mexico City, for example, a World Bank pilot public transport, and gender diversity of scheme has led to the introduction of a mo- bystanders. bile application to make it easier for women In Brazil, the NGO Think Olga created to obtain quick access to the Citizens’ Pro- the collaborative map “Chega de Fiu- tection Unit run by the police. Police officers Fiu”, where users can report and geo-tag from the CPU, together with colleagues sta- occurrences of harassment. tioned along the relevant bus routes, have received training, in partnership with Mex- ico’s Ministry for Women, to provide appro- priate assistance to survivors. 54 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Environmental design to enhance Improving dialogue and collaboration women’s security with the police • Improved lighting31, security cameras, and • Increase police presence in identified police patrols in the communities and stra- hotspots and carry out publicity campaigns tegically placed at bus stops would provide to inform people where and how to report greater protection for women. Interviewees cases of sexual harassment and general vio- commented that their neighborhoods were lence. Given the low levels of trust in the po- calmer when the police were more present. lice in RJ and BA, it is crucial to enhance po- Transport authorities will need to improve lice training and awareness, and introduce collaboration with urban planners to ensure mechanisms to ensure police accountability an integrated approach to enhancing wom- and proper enforcement of the law. en’s security when using transport, and in • Consider allocating more female police of- the urban environment generally. ficers to community policing. In RJ and BA, informants claimed that they distrusted the Capacity building in GBV police. Strengthening contact with female of- responsiveness ficers can enhance women’s sense of safety. • Collaborate with NGOs and government • Training bus operators to respond appropri- authorities responsible for gender issues to ately to sexual harassment of passengers, train police on how to respond to harass- setting up and supporting transport compa- ment and violence against women and girls nies managed by women, and ensuring that in public spaces. government/transport operator concession agreements incorporate Codes of Conduct. Such Codes of Conduct should condemn cas- Training community residents to es of misconduct such as those committed reclaim public spaces and become by staff within the transport company and active agents of change by drivers. The enforcement of these Codes of Conduct defined under the concession • Show people how to intervene in cases of agreements can facilitate the monitoring of sexual harassment and violence in the additional unwanted behaviors beyond sex- communities without putting themselves ual harassment, such as drivers refusing to at risk. Grassroots initiatives as Hollaback! pick up women with bags and children, al- can train communities on bystander inter- leging that they take up space that could be ventions and sexual harassment preven- occupied by paying passengers. tion to reclaim public spaces. Some forms of trained and skilled community patrolling can be put into place for complementary 31 Analysis in refugee camps shows that involving the community in the location of lighting has a significant impact in terms of increased safety and security (UNHCR, 2017). 55 5. RECOMMENDATIONS sure that a clear response route is defined for different kinds of complaints. Ethical BOX 15 considerations should be taken into account Plan International Safer Cities when responding to specific issues posed for Girls program by women where confidentiality should be a key pillar (e.g., on sexual harassment). At The Safer Cities for Girls program of Plan the same time, the response protocol should International promotes the activities “Social ensure that women are referred to the ade- Cartography” and “Safety Walks”. In Lima, “Safe- quate services (e.g., Local Women Institutes). ty Walks” in the District of Carabayllo in Lima allowed young girls to select and visit an area of ​​their community to locate safe and unsafe 5.2. Recommendations to enhance places, accompanied by different stakeholders transport availability from the community such as: community lead- ers, transport operators, authorities, municipal Designing transport services to match police, etc. This data has been shared with demand and regulation local authorities to inform the improvement of infrastructure in the community. • There is a need to analyze the demand by different groups, including women, to plan transport services that address mobility pat- terns and requirements. surveillance, non-confrontational interven- • A formalized transport sector allows up- tion, and civic education. keeping a certain level of quality of trans- port services. Hence the need to design for- • Use computer-gaming devices to promote mal services that address the demand, while safer commuting experiences and invest in providing specific standards for the quality enhancing social capital and community of services through concession agreements participation. Develop applications to facil- and requirements established for autho- itate communication between community rized routes. Formalized services could lead residents so that they can coordinate their to improved schedules, reliable timetables, commutes. Rewards points could be offered regular bus-stops, ensuring that buses stick to people if they arrive at meeting points on to their allotted routes. Service contracts time, or if they report (e.g., with pictures on- can incorporate measures to address wom- line) unsafe infrastructure to the authorities en’s safety and mobility since the beginning. to hold them accountable. Contract provisions can be enforced through • Create effective feedback mechanisms regular inspections and monitoring. where community members can place com- • Public transport “gaps” are often filled by in- plaints and develop a sense of ownership of formal or private (authorized) modes, which the public space. Several transport operators play an essential role in supplementing count with call centers and other feedback women’s mobility requirements. Formaliz- mechanisms where transport users can ing and regulating supplementary transport place charges. Response protocols must en- 56 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES • Public transport reforms are a complicated BOX 16 process involving the need to overcome a multiplicity of barriers. Some of these bar- Improving internal transport riers, such as obtaining access to finance, coverage could be perhaps secured by mobilizing Maximizing Finance for Development to fa- The Urban Transformation Project of cilitate the access by a group of operators to the Buenos Aires Metropolitan Area private loans, or by establishing concession (financed by the World Bank), aimed agreements directly with the government. at integrating Villa 31, a fleet of electric buses will be purchased to enhance internal mobility for residents of the Improving “last mile” connectivity Villa and cover the limited supply in the • Provide alternative short-distance transport, community. The intervention involves such as subsidized bicycle-sharing schemes, reorganizing public areas, introducing and improve conditions for people who pre- an appropriate public transport system fer to walk. Bicycles and walking can make (i.e. buses), restricting the use of pri- it easier and quicker for women to move vate cars, expanding the bicycle rental around in their neighborhoods. Bike-sharing service (Ecobici), improving pedestrian schemes, for example, could encourage wid- footpaths, etc., and facilitating inter- er use of bicycles and lessen the possibility connection with the metro and train of theft of privately-owned bikes. Such pro- stations. The new bus service in Villa grams must be combined with the construc- 31 aims, inter alia, to incorporate the tion of safe bicycle paths and the installation informal “combi” system that currently of bicycle racks on public transport vehicles. serves residents living at the north end of the neighborhood so that they can be • In peripheral communities, the introduction dropped to the nearest transport hub. of better walking facilities, street lighting, and publicity campaigns to accept the idea Meanwhile, in Peru, the World Bank will of women riding bicycles would be an ad- finance a cable car project to benefit vantage. low-income people living on the pe- • Several elements need to be considered for riphery of Lima, and increase connec- bike sharing to be an equalizer for women’s tivity between San Juan de Lurigancho mobility. Analysis from the Urban Institute and the downtown area of the city. in Washington DC showed that there are few bike racks in low-income areas and that those that do exist are more likely to be near to transport hubs rather than to people’s services (unauthorized minivans, taxis, and homes. This entails people resorting to other moto-taxis) can be a crucial initiative for in- modes of transport to access bike racks, thus creasing the safety of women travelers and undermining the potential for bicycles to im- ensure they are not subject to price discrim- prove mobility for the first and last mile of ination. journeys. Information on cheap bicycle rental 57 5. RECOMMENDATIONS BOX 17 Positive effects of encouraging more women to cycle A recent study for the Women for Climate Initiative conducted by the University of Berke- ley, California, showed that encouraging more women to take up cycling can contribute directly to carbon emissions reduction. To promote urban cycling, the San Francisco “Transit First Policy” acknowledges the importance of material infrastructure (e.g. pro- tected bike lanes) and of investments in sociocultural infrastructure (e.g. working with the community to change gender bias towards bicycle users in order to increase bike use by women). Another study (by the University of Kentucky) done in some large US cities (Grae- hler et.al., 2019) shows a link between cycling and increased use of the metro and light- rail, while reduced bus ridership was noted to have a positive impact on carbon emissions reduction. In Lima, a new law was approved on April 24, 20191, aimed at introducing specific mea- sures to encourage and regulate bicycle use in Peru as an efficient and environmentally sustainable mode of transport. This initiative should also have a positive impact on im- proving women’s access to bicycles. Rio’s Transport Department is currently developing a 6-month pilot project at a suburban (SuperVia) rail station to provide bicycle racks and bike rentals on weekday and weekends to local residents. This facility is expected to improve commuting for residents who live near the station by improving “last mile” connectivity (multi-mode integration). Rio’s Su- perVia suburban rail system caters for the city’s poorest inhabitants, with102 stations and a 270 km rail network serving some of Rio’s most dangerous neighborhoods. From a policy perspective, it could also be explored the possibility of encouraging the use of bicycles by children to go to school. This kind of initiative would have to be comple- mented with safe routes that include programs that go from safe street crossings, main- taining sidewalks, to education programs for students and the communities on how to bike safely, among others. 32 32 https://busquedas.elperuano.pe/normaslegales/ley-que-promueve-y-regula-el-uso-de-la-bicicleta-como-medio- ley-n-30936-1762977-4/ 58 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES rates should be disseminated in spaces fre- quently accessed by women, while flexible BOX 18 payment methods should be considered for low-income people without credit cards. Planning for inclusive cities: the Given the credit accessibility gender gap (i.e., case of Medellin women have less access to credit), this could Medellin represents an example of a help boost women’s mobility. city that, through adequate transport and urban planning, promoted inclu- Integrated transport systems siveness and contributed to address its situation of criminality and violence. • Enhancing institutional capacity to ensure As part of this inclusive planning, the the provision of sustainable urban trans- public transport network and physical port involves improving coordination and street infrastructure layout were de- integration between transport services. In- veloped in a way to increase access to tegration must not be restricted to tariffs. significantly better economic opportu- It should also include actual physical inte- nities for the people living in the slums gration, and operational integration, such as in Medellin. Public transportation better coordination between different ser- (metro, metrobus, cable car) was highly vices at night aimed at reducing the risks to integrated with pedestrian and cycling women enduring long waits in bus stations infrastructure. Public spaces were or at bus stops. In Peru, the law governing also transformed into safe and livable the creation of the Lima and Callao Urban neighborhoods that included lighted Transport Authority (ATU), published on De- streets, playgrounds, public libraries, cember 28, 2018, aims to organize, imple- waiting areas, ramps, etc. This urban ment and manage the Integrated Lima and transformation was complemented by Callao Transport System. The creation of the new social, educational, cultural and ATU, which heralds a complete transforma- violence prevention programs (Ijjasz- tion of transport in the Peruvian capital, will Vasquez and Duran Vinueza, 2017). All involve robust institutional and inter-insti- this enhanced social cohesion and the tutional strengthening. sense of community belonging that led to violence prevention. Planning for inclusive urban spaces • Urban spaces can be designed for promoting inclusion and boosting women’s capabilities. Cities are inclusive when they equally provide ing connectivity and accessibility to main in- access to services and economic opportunities dustrial, commercial, and employment areas. to different population groups. Cities that plan Easing women’s access to economic opportu- from an inclusive approach think since the nities and services can have a direct impact on outset on increasing people’s mobility, provid- women’s economic empowerment; thus, their ing neighborhoods with services like childcare, access to resources and decision making with- health, and education facilities, plus improv- in their households. 59 5. RECOMMENDATIONS Regulatory restrictions that may 5.3. Recommendations to enhance create barriers for the poor transport affordability Excessive regulation (e.g., maximum floor area Fare scheme suited to women ration, minimum lot size, forbidden mixed uses) mobility patterns can limit the supply of built space, raising pric- es artificially and excluding the poor. Thus, an • Analyzing the mobility patterns for different assessment of urban regulation must be taken socio-economic groups and designing inte- into account in urban planning, so that low-in- grated fare schemes can reduce multimodal come populations are not socially expelled to travel costs and enable the introduction of distant, disperse, and disconnected areas, as targeted subsidies. The design and imple- they cannot afford prices in more centrical and mentation of such policies must be found- denser ones. Better regulation, on the contrary, ed on a financially sustainable, balanced can promote, for instance, mixed usage of land approach aimed at improving transport ef- (including social housing) that will increase ficiency while recognizing the affordability supply and reduce prices leading to more inclu- constraints suffered by low-income women. sive and better-designed cities. Some integrated fare schemes could involve, for example, exemption from “transfer” costs and the introduction of cheaper daily, week- ly, or monthly tickets. BOX 19 Transport subsidies for the poor in Buenos Aires and integrated fare in Sao Paulo Demand-side subsidies were launched in 2014 to protect poor people from fare increas- es. With the SUBE personalized smartcard, cheaper “social” tariffs can be bought by the city’s most vulnerable citizens [1]. The discount has increased recently from 40% to 55% of the full fare, and 1 million additional people now benefit from the scheme (2.3 million in December 2018 in the Buenos Aires Metropolitan Area. In early 2018, the Argentine Government also introduced an integrated fare system for public transport in the AMBA, providing a discount of 50% on the second trip and 75% on further trips. These subsidies benefit the poorest users given that they tend to undertake longer trips involving more transfers (World Bank study, 2018). This new fare structure is better suited to address the needs of low-income women in particular, who live in pe- ripheral areas, carry a larger burden of the household travel needs and bear larger travel complexity with multi-stop trips. 60 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Average number of transfers per trip +3% 1.4 1.2 1 0.8 +25% 0.6 0.4 0.2 0 All trips with full fare All trips with social tarife Trips with transfers and full fare Trips with transfers and social tarife [1] Beneficiaries of Asignación Universal por Hijo, Programa de Jefes de Hogar, Asignación por embarazo, Plan Progresar, Personal de Trabajo Doméstico, Programa Argentina Traba- ja y Ellas Hacen, Monotributo social, and retirees and pensioners. BOX 20 Integrated fare in Sao Paulo, Brazil Bilhete Único is the São Paulo transportation contactless smart card system for user fare control. It is managed by SPTrans and integrates metro, buses (onibus) and suburban railways. Special fares apply to (i) students and teachers who pay half price; (ii) people with disabilities and the elderly (women over 60 years old and men over 65 years old) are ex- empt from payment; (iii) pregnant women have an special card called “Bilhete Único Mãe Paulistana“. This card guarantees free transportation to pregnant women registered in the UBS - Basic Health Units and benefited by the Paulistana Mother Program of the Municipal Health Secretariat. There is also a special card called “Bilhete Único Amigão” which pro- vides an special fare valid on Sundays and holidays that allows users to make up to 4 bus trips in the period of 8 hours, at the cost of only one fare; and (v) “Bilhete Único Vale-Trans- porte”. The transportation voucher is a labor benefit under federal law, and it is paid by the employer. The discount on the worker’s salary is limited to 6%. 61 5. RECOMMENDATIONS BOX 21 Improving women’s mobility through childcare provision Childcare services provided by the government could be a good option to incentivize wom- en’ return to employment after childbirth, as evidence shows that they are the primary caregivers of children (WBL, 2018). A study showcased that, among OECD economies, the availability of public childcare for children below the age of 5 years is strongly correlated with employment rates of mothers with young children (OECD, 2001). Another option to ease women’s double burden could be the provision of personal income tax deductions for childcare. According to the WB Women Business and the Law 2018 Report, 33 of the cov- ered economies provide this kind of incentive. In Colombia, the Programa de Hogares Comunitarios de Bienestar has been highly recog- nized for its full country coverage and for benefitting children below 5 years old in margin- alized communities in urban and rural areas. The Program is organized by the beneficiary families: “communitarian mothers” receive in their homes around 12 to 14 children up to five years old and provide them with care, initial education and nutrition. Communitari- an mothers receive a stipend that varies depending on the number of children (González Ramírez and Durán, 2012). More evidence is needed to showcase the impact of childcare provision on women’s mobil- ity and access to better economic opportunities, including an analysis of the best location (e.g. community vs close to train stations) and required regulations for these services, as increasing women’s mobility could compromise the child’s quality of care. Moreover, child- care provision shall be complemented by efforts to change social norms so that co-respon- sibility can happen within the household. 5.4. Additional recommendations ment Infrastructure. A starting point could to enhance women’s transport be to design a pilot project to identify suit- able locations for daycare facilities. Another mobility and accessibility option to be piloted could be the provision Notwithstanding the influence of traditional of adequate breastfeeding facilities in trans- social norms, we identify a set of measures that port hubs. may help to alleviate this burden in short to • Create extracurricular activities in safe plac- medium term: es for older children and adolescents after • Expand the number of vacancies and nearby the end of classes. This can contribute to bet- options for daycare units and kindergartens ter coordination of entry and exit times and for younger children, or include daycare fa- can serve to reduce the number of trips that cilities within a Transport Oriented Develop- mothers have to take to and from school. In 62 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES BA, women’s concern with their children’s BOX 22 involvement in crime was a key reason why they chose not to work away from home. Implementation of methods to improve work/family balance in • Engage community leaders to organize households by changing gender mutual support groups with the use of norms WhatsApp, etc. These groups could encour- age women to help with care activities and Promundo is a Brazilian non-gov- facilitate joint commutes. In BA, we found ernmental organization (NGO) which that WhatsApp groups made women’s care promotes men’s caregiving and active duties easier to arrange. In RJ and LM, joint fatherhood with a view to fostering commutes engendered a sense of safety equitable gender roles, preventing among women who traveled to work. violence against women and children, • Establish school transport systems or “safe and contributing to positive maternal corridors” for schoolchildren to go safely to and child health outcomes. The NGO and from school or to use bus stops (accom- collaborates with a network of NGOs panied or not) on the periphery of the com- around the world that offer training munity. These spatial measures should not courses, particularly for new fathers exclude designing and implementing other and couples. These courses have been security measures in dangerous areas. Im- followed by more than 250,000 people. proving community mobility exclusively in SASA! is a preventive approach to one area can lead to avoided areas becoming violence against women aiming at even more unsafe. addressing power relations as the root • Work with NGOs to raise awareness in com- of gender inequality. The interven- munities of the need to encourage couples tion walks communities step-by-step to share domestic chores (i.e., new roles for through a process of change, and it men) and develop more positive attitudes to- avoids the chronic cycle of aware- wards women’s paid work. ness-raising. The process involves all • Use transport design in infrastructure to members of the community, working promote equal gender norms: In the city on a personal level to ensure a trans- of Vienna, metro wagons have a designated formational approach. SASA! has been space for strollers, wheelchairs, or people rigorously evaluated, and the results traveling with packages. These areas are ac- showcased a significant reduction in companied by combined images of women social acceptance in Intimate Partner and men holding babies distributed random- Violence (IPV) among women and men ly along said wagons to convey the message and lower levels of past year experience that childcare should be seen as a combined of sexual IPV. duty between women and men. 63 5. RECOMMENDATIONS Boosting self-efficacy and self-confidence • Promote mentoring programs: Identify wom- en in the community who are more used to commuting downtown to work. Arrange for these to accompany less confident women and show them how to use public transport. • Capacity building: Design workshops where women can learn skills to search for infor- mation about employment opportunities, apply for jobs, and conduct themselves pos- itively at job interviews. • Empower women in the poorest areas to get used to using public transport by enabling women to assert greater control over their mobility and lives. For example, women can learn how to handle complaint mechanisms in public transport, how to activate service quality and safety feedback. An inspiring ex- ample is the Bike Anjo33 program in São Paulo, which involves volunteer cyclists accompa- nying novices to obtain traffic information, find the best routes to get around the city, etc. 33 https://bikeanjo.org/ 64 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 6 CONCLUDING REMARKS 65 6. CONCLUDING REMARKS This study explored the constraints and facilitators to women’s agency in mobility and work in low-income settings of three Metropolitan Regions in Argentina, Bra- zil, and Peru. While low-income populations are generally the most disadvantaged by deficient public transport services, women are particularly affected. The study identi- fied several constraints on women’s mobility. Some of these constraints are of a more practical nature, such as the availability (or not) of childcare services, while others are subjective and relate more to the social context – values, norms, and structures. The various constraints identified by this study shape women’s mobility preferences and choices in terms of the transport they use. Findings showcase how transport gaps overlap date their associated care responsibilities. This with poverty and gender-based constraints, pattern contrasts sharply with men’s work-re- restricting the opportunities (poorer) women lated preferences and decisions, which are cen- can realistically take up to advance their lives. tered on improving their job status and salaries, This contributes to a powerful vicious cycle of regardless of distances. disadvantage. Gendered mobility constraints The interaction between safety, affordabil- represent a more substantial burden for wom- ity, gender norms and internal constraints en—in terms of time, cost, and wellbeing—in shapes women’s mobility preferences differ- comparison to men, shaping their decisions ently to men’s. In areas with high levels of crime and advancement in their professional lives and violence, female commuters prioritize safer differently. Women who work outside the home travel options whenever they can afford them, experience constant exposure to safety risks, often implying longer travel times and lower fear, and stress while commuting, which harms disposable income at the end of the month. For their emotional and psychological wellbeing. example, men in this study commonly made Transport deficiencies add to the list of fac- segments of their trip through walking (cheap- tors that discourage women from joining the er) or moto-taxis (faster), while these options wider labor market, forcing them into part- were considered too risky for most women. Im- time, low-wage jobs nearer home, or encour- plicitly, these women become subject to a “pink aging them to opt out of paid work altogether. tax,” which men don’t have to pay, as they feel At the household level, inequality in the alloca- comfortable using modes which are normally tion of duties shapes mobility and work deci- not an option for women. The study also reveals sions. We find that, for example, when looking the situation of a normally forgotten group of for a job, women will prefer job proximity over women with overlapping vulnerabilities (such job quality34, taking up opportunities that are as young single mothers with less affordabili- informal but that are closer to home and have ty capacity). These women cannot pursue their flexible hours. This relieves the burden of com- preferences for safer transport and will have muting as well as enables them to accommo- to settle for cheaper options in detriment of 34 “Higher quality” jobs are those with higher productivity, better pay and working conditions (World Bank, 2012a). 66 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES their wellbeing. Zooming into the heterogeneity the micro level for each site does not allow us to among women makes evident the need to com- reliably compare the socioeconomic profile of plement policies that tackle sexual harassment high and low accessibility areas for each met- and personal security (in public transport and ropolitan region. However, the data collected communities more broadly) with more afford- for this study suggest that—apart from a more able tariff schemes. constrained set of jobs to access—low accessi- bility areas are also characterized by lower in- The findings also show that safety and af- comes, fewer transport options, less access to fordability constraints are exacerbated by un- social services, and, in most cases, higher levels available transport options in both low and of crime and violence. Women in these loca- high accessibility areas. Given the unavailabil- tions face the most significant burdens as they ity or unreliability of their preferred transport have to commute for longer times and distanc- options (especially during off-peak hours and es, are exposed to more risk, and have to spend the night), women are forced to use informal significantly more from their own pockets to be services they deem risky but that are the only able to engage in paid work. ones that do cover the last segment of their trips (e.g., motorbikes). A key conclusion from our findings is that improving infrastructure alone is not sufficient Women mitigate the risks they face by de- to ensure the take-up of transport services by veloping coping strategies and relying on their women. If women’s preferences, constraints, social networks. Our study shows that the and choices are not taken seriously into ac- stronger the social networks are, the easier it is count, transport interventions will not succeed for women to accept jobs outside of their com- in catering for the most vulnerable women and munities. In the absence of formal childcare fa- are, moreover, unlikely to succeed in promoting cilities, technology (e.g., WhatsApp groups) can gender equality. While evidence indicates that play an important role in facilitating mutual other priorities concerning the mobility of the support within the community for household general population can often eclipse the issues and care activities. Women also use strategies affecting women, there are many opportuni- to improve their sense of safety while commut- ties for including gender-responsive interven- ing, for example, asking their partners or family tions in infrastructure and transport projects, members to accompany them to bus stops. Al- not only through infrastructure improvements though this support is helpful, it can gradually but also complementary interventions. Some of undermine women’s sense of autonomy and these interventions are low hanging fruits that self-efficacy for independent travel. can easily be incorporated into the planning The study finds the same patterns in high and design of transport services from the out- and low accessibility areas, although trans- set (e.g., design features that address differenc- port burdens weigh more heavily on women es in mobility patterns and campaigns). Anoth- living in low accessibility areas. This relates er set of more transformational interventions to the greater physical distance between their will require inter-governmental collaboration, place of residence and job opportunities, but sensitization of stakeholders (e.g., available can also be attributed to their exposure to ad- childcare options and design of community in- ditional vulnerabilities. The lack of statistics at frastructure), and bringing civil society organi- 67 6. CONCLUDING REMARKS zations on board (e.g., for bystanders and social norms activities) in order to be sustainable. The intervention threshold will depend on the ca- pacity, political will, and availability of resourc- es of the involved parts. Developing more and better-designed pilots and analytics is key to demonstrating that transport can contribute to gender equality while, at the same time, im- proving the effectiveness and inclusiveness of transport policies. 68 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES REFERENCES Balassiano, R. (1998) “Planejamento estratégico tunities and Challenges. London: Develop- de transportes considerando sistemas de ment Planning Unit University College média e baixa capacidade”. Transporte em London. Transformação II, Ch. 9, Makron Books, São CAF, 2016. Observatorio de Movilidad Urbana: In- Paulo. forme 2015–2016. “Urban Mobility Indica- Balassiano, R. and Alexandre, R. (2013). Informal tors for Buenos Aires, 2015.” https://www. Motorized Transport in Rio de Janeiro, Brazil. caf.com/es/temas/o/observatorio-de-mo- Case study prepared for Global Report on vilidad-urbana/bases-de-datos/. Human Settlements. Ciclocidade. (2016). “Perfil de quem usa bicicle- Berkeley SafeTREC (2017). City of Monterey Safe ta na cidade de São Paulo”. Accessed at: Routes to School Action Plan, California: https://www.ciclocidade.org.br/bibliote- Berkeley. ca/pesquisa-ciclocidade/file/113-relato- rio-completo-pesquisa-perfil-de-quem- Bhide, Amita (2016). “Engendering Mumbai’s usa-bicicleta-na-cidade-de-sao-paulo Suburban Railway System”, Centre for Urban Policy and Governance, Tata Insti- CIVITAS (2014). “Smart choices for cities: Gen- tute of Social Sciences. der Equality and Mobility, Mind the Gap!”, CIVITAS, 2020. Blanco, J., & Apaolaza, R. (2018). “Socio-territo- rial inequality and differential mobility. Scovino, Adriana (2008). “As viagens a pé na ci- Three key issues in the Buenos Aires Met- dade do Rio de Janeiro: um estudo da mobili- ropolitan Region”. Journal of Transport Ge- dade e exclusão social”. Dissertação de M.Sc. ography, 67, 76-84. Programa de Engenharia de Transportes. COPPE/UFRJ. Rio de Janeiro, Brasil Blumen, O., and A. Kellerman. 1990. “Gender differences in commuting distance, resi- Cristaldi, F. (2005). “Commuting and gender in dence, and employment location: Metro- Italy: a methodological issue”. The Profes- politan Haifa”, 1972–1983. The Professional sional Geographer, 57(2), 268-284. Geographer 42: 54 – 71. Dominguez, K., Kumar S. & Mehndiratta S. Bocarejo S., J. P., & Oviedo H., D. R. (2012). “Trans- (2015). When good transport alone doesn’t port accessibility and social inequities: a bring jobs closer to women: insights from tool for identification of mobility needs and Mexico City. Washington, D.C: World Bank evaluation of transport investments.” Jour- Group. nal of Transport Geography, 24, 142–154. Doyle, D. G., & Taylor, B. D. (2000). Variation in https://doi.org/10.1016/j.jtrangeo.2011. Metropolitan Travel Behavior by Sex and 12.004 Ethnicity. In Travel Patterns of People of Cañón-Rubiano, Leonardo. (2010). Transport and Color (pp. 181–244). Washington, D.C.: U.S. Social Exclusion in Medellín. Potential Oppor- Department of Transportation. 69 REFERENCES Duchene 2011, Chantal. “Gender and Transport”, Haeran, Shin. (2010). “Spatial Capability for Discussion paper International Transport Understanding Gendered Mobility for Forum, OECD, 2011. https://www.itf-oecd. Korean Christian Immigrant Women in org/sites/default/files/docs/dp201111.pdf Los Angeles”. London: University College London. Elias, W., Newmark, G. L., & Shiftan, Y. (2008). “Gender and travel behavior in two Arab Hanson, S. (2003). “Geographical and feminist communities in Israel”. Transportation perspectives on entrepreneurship”. Geog- Research Record, 2067(1), 75-83. raphische Zeitschrift, 1-23. Lima Como Vamos. (2017). Encuesta Lima Como Hanson, S. (2010) “Gender and mobility: new Vamos. VII Informe de percepción sobre approaches for informing sustainability”, calidad de la vida en Lima y Callao. Lima: Gender, Place & Culture, 17:1, 5-23, DOI: Lima Como Vamos. 10.1080/09663690903498225 Encuesta de Movilidad Domicialiaria: Movilidad Ijjasz-Vasquez E. & Duran Vinueza S. (2017). del área Metropolitana de Buenos Aires. “How is Medellin a model of urban trans- 2010. Buenos Aires. formation and social resilience?” Wash- ington, D.C.: World Bank Group. Fernando, D., Cohen, L., & Duberley, J. (2018). What managers can do to keep women in Índice de Bem Estar Urbano (IBEU) Rio de Janei- engineering. Harvard Business Review. ro - Observatório das Metrópoles. http:// www.observatoriodasmetropoles.net/ Geurs, K.T, Van Wee, B., (2004). “Accessibili- index.php?option=com_k2&view=item ty evaluation of land-use and transport &id=665:mobilidade-%C3%A9-a-pior-di- strategies: review and research direc- mens%C3%A3o-do-bem-estar-urbano- tions”. Journal of Transport Geography 12, no-rio-de-janeiro&Itemid=164&lang=pt# 127–140. ILO (2017). World Employment and Social Outlook: Gomide, A., Leite, S., & Rebelo, J. (2005). Public Trends for women 2017. International La- transport and urban poverty: a synthetic bour Office – Geneva: ILO, 2017. index of adequate service. Documento de Trabajo, (33901). ITRANS. (2003). MOBILIDADE E POBREZA: Região Metropolitana do Rio de Janeiro. Gossen, R., & Purvis, C. L. (2005). Activities, time, ITRANS, 2003. Link: http://mcadaval.com. and travel: changes in women’s travel br/blogs/relatorio_RMRJ.pdf time expenditures, 1990–2000. Research of Women’s Issues in Transportation, 2. Kondylis, F., Legovini, A., Vyborny, K., Zwager, A., Andrade, L. (2019). Demand for “Safe Graehler, M., Mucci, A., & Erhardt, G. D. (2019). Spaces”: Avoiding Harassment and Stig- Understanding the Recent Transit Rider- ma. Washington, D.C: World Bank Group ship Decline in Major US Cities: Service Cuts or Emerging Modes?. In Transporta- Koch, J., Lindau, L. A., & Nassi, C. D. (2013). Trans- tion Research Board 98th Annual Meet- portation in the Favelas of Rio de Janeiro. ing, Washington, DC, January. Lincoln Institute (www. lincolninst. edu). 70 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Lecompte, M.C and Bocarejo, JP. (2017). “Trans- Making Inroads for Women: A Qualitative port Systems and their Impact on Gender Study on Constraints and Opportunities of Equity”, Transportation Research Procedia, Women’s Equal Participation in the Roads Sec- Volume 25, 4245-4257. tor in Malawi. Washington, DC: World Bank Group https://openknowledge.worldbank. Lee, B. S., & McDonald, J. F. (2003). “Determinants org/handle/10986/31656 License: CC BY of commuting time and distance for Seoul 3.0 IGO. residents: The impact of family status on the commuting of women. Urban Studies, Noack, Eva. (2010). “Stuck in the countryside? 40 (7), 1283-1302. Women’s transport mobility in rural Ab- erdeenshire, Scotland – experiences, be- Litman, T. (2017). Evaluating Accessibility for haviour and needs.” Vienna: Facultas Ver- Transportation Planning: Measuring Peo- lags. ple’s Ability to Reach Desired Goods and Activities. Transportation Research (p. 62). Organisation for Economic Co-operation and Development (OECD). (2001). “Balancing Loukaitou-Sideris, A., & Fink, C. (2009). Ad- Work and Family Life: Helping Parents dressing women’s fear of victimization into Paid Employment”. OECD Employment in transportation settings: A survey of US Outlook 2001 (Ch.4). Paris: OECD. transit agencies. Urban Affairs Review, 44 (4), 554-587. Peralta, T., Mehendiratta S. & Ochoa C. (2014). Gender, Travel and Job Access: evidence Marion, B., & Horner, M. W. (2007). Comparison from Buenos Aires. Washington, D.C.: of socioeconomic and demographic pro- World Bank Group. files of extreme commuters in several US metropolitan statistical areas. Transporta- Peralta, T. (2015). “Mobility for All: Getting the tion Research Record, 2013(1), 38-45. Right Urban Indicator Shifting from the Proximity of Transport to the Accessi- McGuckin, N., & Nakamoto, Y. (2005). Differenc- bility of Opportunities”. Transport Brief. es in trip chaining by men and women. Washington, D.C.: World Bank Group. In Conference proceedings (Vol. 35, pp. 49-56). Perova, E., & Vakis, R. (2013). Improving Gen- der and development outcomes through McLafferty, S., & Preston, V. (1991). Gender, race, Agency: policy Lessons from Three pe- and commuting among service sector ruvian Experiences. Lima: World Bank. workers. The Professional Geographer, https://openknowledge.worldbank.org/ 43(1), 1-15. handle/10986/16259 License: CC BY 3.0 Murakami, E., & Jennifer, Y. (1997). Daily Travel by IGO.” Persons with Low Income Daily Travel by Peters, Deike. (2002). “Breadwinners, Homemak- Persons with Low Income. African Amer- ers and Beasts of Burden: a Gender Per- ican Mobility Symposium, 1–27. https:// spective on Transport and Mobility”. Insti- doi.org/10.1111/j.1365-2826.2006.01429.x tute for City and Regional Planning (ISR), Müller, Miriam, Sevara Melibaeva, Ana Luiza Berlin, Germany. Machado and Ursula Casabonne. 2019. 71 REFERENCES Pero, V., & Mihessen, V. (2013). “Mobilidade ur- Impact of Gender Planning in Europe. bana e pobreza no Rio de Janeiro”. Revista London: Ashgate. Econômica, 15(2). Sant’Anna, R.M. and R. Balassiano (2000) “Or- Plan International. (2018). “Creating safer cit- ganização funcional do transporte al- ies for and with girls”. United Kingdom: ternativo: Evolução e possíveis cenarios https://plan-international.org/ending-vi- futuros”, Engenharia de Tráfego e Trans- olence/safer-cities-girls portes 2000: Avanços para uma Era de Mudanças, XI Congresso Panamericano Polk, M. (2003). “Are women potentially more de Engenharia de Trânsito e Transporte, accommodating than men to a sustain- November 2000, Gramado, pp.415–428. able transportation system in Sweden?” Transportation Research Part D: Trans- González Ramírez, JL, & Durán, M. (2012). Eval- port and Environment, 8(2), 75-95. uation for Improving: The Impact Eval- uation to the Program Hogares Comuni- Pucher, J., & Buehler, R. (2008). Making cycling tarios de Bienestar del ICBF: http://www. irresistible: lessons from the Nether- scielo.org.co/scielo.php?script=sci_art- lands, Denmark and Germany. Transport text&pid=S0120-35842012000100007 reviews, 28(4), 495-528. Shaw, M. (2002). “Women’s Safety: Is there Prog- Pulido, D., Darido, G., Munoz-Raskin, R., & ress in Putting Gender into Crime Pre- Moody, J. (Eds.). (2018). The urban rail de- vention?”. First International Seminar on velopment handbook. The World Bank. Women’s Safety – Making the Links, Mon- Raising Voices. (2019). SASA! Mobilizing Com- treal, Canada, May 2002. munities to Inspire Social Change: Srinivasan, S. (2008). A spatial exploration of http://raisingvoices.org/wp-content/up- the accessibility of low-income women: loads/2013/03/downloads/resources/Un- Chengdu, China and Chennai, India. Gen- packing_Sasa!.pdf dered mobilities, 143-158. Rosenbloom, S. 2006. “Understanding women’s Tanzarn, N. (2008), “Gendered mobilities in de- and men’s travel patterns: The research veloping countries: the case of (urban) challenge”. In Research on women’s is- Uganda”, in T. P. Uteng and T. Cresswell sues in transportation: Volume 1 Con- (eds), 159–72. ference overview and plenary papers, Conference proceedings 35, 7 – 28. Wash- UN Habitat (2018). Revision of World Urbaniza- ington, DC: National Research Council. tion Prospects. Department of Economic Affairs. New York: United Nations. Root, A., Schintler, L., & Button, K. (2000). Women, travel and the idea of’sustainable trans- UN Women. (2013). Study on the Ways and port’. Transport Reviews, 20(3), 369-383. Methods to Eliminate Sexual Harassment in Egypt, UN Women, 2013. Sánchez de Madariaga, I. (2013). “The Mobility of Care: A new Concept in Urban Trans- UN Habitat. (2008). Women’s safety audits. portation”. In Sánchez de Madariaga, I., What works and where? New York: Unit- & Roberts, M. (Eds.), Fair Share Cities: The ed Nations. 72 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Uteng, Tanu Priya. “Gender and Mobility in the World Bank. (2012b). World development report Developing World: Gendered Bargains of 2012: Gender equality and development. Daily Mobility”, World Development Re- World Bank Publications. port, 2012. World Bank Group, IDB, ICRW (2015). “Violence Vance, C., & Iovanna, R. (2007). “Gender and the Against Women and Girls”, Transport Brief. automobile: analysis of nonwork service World Bank, 2015. Peru - Lima Metro Line Two trips”. Transportation Research Record, Project. Project Appraisal Document. 2013(1), 54-61. World Bank. Independent Evaluation Group. Vandermissen, M., M. Theriault, and P. Ville- (2017). Argentina: Buenos Aires Urban neuve. 2006. Work trips: Are there still gen- Transport Project. Project Performance der differences? The case of the Quebec Assessment Report. metropolitan area. Paper presented at the Transportation Research Board Annual World Health Organization. (2005). Researching Meeting, January 22 – 26, Washington, DC. violence against women: Practical guide- lines for researchers and activists. World World Bank (2010). Mainstreaming Gender in Health Organization. Road Transport: Operational Guidance for World Bank Staff, Transport Papers, 2010. World Health Organization. (2012). Under- standing and addressing violence against World Bank. (2012a). World development report women: Intimate partner violence. WHO/ 2013: Jobs. World Bank Publications. RHR/12.36. World Health Organization, 2012. 73 ANNEX ANNEX 1: RESEARCH TOOL QUALITATIVE RESEARCH: to help me remember what we have talked about. The notes will be useful for asking you URBAN MOBILITY IN LATIN AMERICA about anything later. Is that okay with you?” IN-DEPTH INTERVIEW GUIDE FOR USE 4. When you start the recording, keep the in- WITH INDIVIDUALS terviewee calm by reminding him/her of the aim of the interview: 1. Interviewer: please follow the following “I would like to remind you that the purpose of protocol (standard procedure) at the begin- our conversation will be about your day-to-day ning of the interview: life, the difficulties you have with transport and Read and sign the Terms of Consent: “Before your opinions on the different modes of trans- we start, I want to read this term of consent port that you currently use. There is no right or to you. It is a standard document used in this wrong answer. We simply want to know about type of research. It sets out my responsibilities your own experiences. Here is a questionnaire as an interviewer. It should help to clear up any containing a list of questions that are asked to queries before the start of the interview”. Inter- all the interviewees. The idea is for me to ask viewer: Read the Terms of Consent to the in- you a question and you will simply give your terviewee. After doing this, clear up any doubts answer. The questions are very straightforward, he/she might have. Sign both copies of the ToC. okay? Hand one to the interviewee and keep the other 5. During the interview, encourage the inter- for yourself. viewee to provide more information, using 2. Tell the interviewee again how long the in- phrases such as: terview will last. “You are the one who really knows”, “You are 3. Ask for permission to use a recorder: the expert” “Can we start? May I use a recorder? It’s just so 6. IMPORTANT. When you start the recording, I can remember what we have talked about. I give the interview identification number. can stop the recording at any time. If you want Do not mention the full name of the inter- to talk about something without recording it, let viewee. me know, and I will pause the recording, okay? (INTERVIEWER: Cover all the questions, but not During our talk I might write down a few notes necessarily in the order below). 74 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES e. Number, age, and gender of siblings? MANDATORY INFORMATION: f. With whom do you live? • - Interviewee’s age g. Do you have a spouse or partner? Do you live with him/her? If not, where does your • - If he/she has children (how many partner live? How long have you lived in and how old are they) this community? How do you feel about • - Schooling (last level) this community? Do you intend to move • - Marital status out? Why (if yes/no)? 2. How do you share domestic responsibilities at home? The interview’s objectives are: a. What is your house like, and how is the • Explore the interviewee’s mobility patterns; space arranged? How many rooms do you have, how many bathrooms, who sleeps • Explore the objective and subjective factors where? that enable or restrict the interviewee’s mo- bility; b. How do you organize household chores? For example, who cooks? Who cleans the • Explore the experiences and perceptions house? Who does the laundry? [i.e., in- about the interviewee’s and his/her part- terviewer trying to find out about the di- ner’s work. vision of tasks between generations and • Explore how both objective and subjective genders]. mobility factors may facilitate/impede ac- c. How is your relationship with your family cess to job opportunities. or the people you live with? d. What do you think about the division of PART 1: Getting to know the tasks in your household? interviewee’s and his/her family e. How do you think this division of labor af- members’ profiles and relationships fects your life? 1. I want you to tell me a little more about 3. Now I want to talk about relationships. Can your family members - your parents, sib- you tell me how you met your last partner? lings, other close relatives, children, with [This question will vary depending on the an- whom you live at present. swer: If the interviewee does not have a partner/ spouse, ask about his/her latest or most import- a. Where were you born? Where did you live ant relationship]. during your childhood, and with whom? a. When did you both meet, and how old b. Your parents’ place of birth? were you? c. Migration History/Reasons b. Were you or your partner married before? d. Do you have other relatives in the neigh- Did your partner already have children? borhood? c. What was your relationship like? 75 ANNEX d. If you have children - children’s birth cir- ions. First, I want you to tell me about an cumstances, whether or not they were ordinary day in your life, from Monday to planned? Friday, from the moment you wake up until you go to sleep. Please give me as many de- e. If you do not have children, are you think- tails as possible. For example, what is your ing about having any? schedule? What do you usually do in the f. What were the circumstances that morning, afternoon, and evening? Who do brought the relationship to an end? you typically spend the day with? How do 4. Circle of relationships. Now, I would like you organize yourself to go out? Who usual- you to tell me who the people are that you ly goes with you? can rely on, where they live, and how they a. How do you spend your time and in what communicate with you — [Explain to the in- kind of activities? terviewee that these people can be in different b. How do you organize the activities in the places]. schedule you mentioned? a. For example, do you have people that you c. What do you do at weekends? can rely on for everyday things, such as if you need to borrow something from d. What was different when you were single them? What kind of support do you have? or had no children? Please tell me about Where do these other people live? How other aspects, such as a relative’s health often are you in touch with them? What problems or other tasks that occupy your channels of communication do you use time. (face-to-face, social media, etc.)? 6. DAILY EXPERIENCES: Based on what you b. Do you have people that you can rely on have told me, I would like to know a little for more serious things or people to talk to more about your day-to-day experiences. when needed? What kind of support can I’d like you to tell me about: 1) your every- they provide? Where do these people live? day journeys (work, school, medical appoint- How often are you in touch with them? ments, visits to relatives’ houses, shopping What channels of communication do you (and anything else you can mention), 2) the use (face-to-face, social media, etc.)? modes of transport that you currently use (walking, colectivo, train, metro/subway); 3) c. Do you know people that count on you how long does the journey take; 4) how do for help? What kind of support? Where do you feel during your journey, from depar- they live? How often are you in touch with ture to arrival - what is good and what is bad them? Form and channel of communica- about your experience? tion (face-to-face, social media, etc.)? [For each journey, ask] “ to get to “X” (place PART 2: Mobility experiences and mentioned)” the following questions: perceptions a. What means of transport do you use? b. What would be an alternative means of 5. In this second part, I would like to talk transport, and which one do you prefer? about your transport experiences and opin- 76 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES c. What days/hours do you usually travel? a. Who has an income-generating activity in What would be other transport alterna- your household? Formal or informal? tives, and why do you prefer any particu- b. Does anyone in your household go out to lar one? work? Who? What do you do and where? d. If applicable: What is the journey like? (in- [Father/mother/siblings, for married in- frastructure, bus stop, etc.). terviewee - ask about spouse or partner] e. How long does it take to you to get to your c. At present, do you have your own income? destination (“X”) from home (or another Do you earn this money by yourself, or starting point), including getting to the are third-party sources involved, such as stop and waiting for transport? a monthly stipend, a retirement pension, social assistance benefits, financial sup- f. What is it positive, and what is it negative port from your spouse? about this journey? [Important: Pose it as an open question combining both polarities. Let d. What is the relative weight of these earn- the participant decide to start with negative or ings in overall family finances, i.e., whose positive]. money, and in what proportion, is used to cover different kinds of expenditure?). • Perceptions of comfort, price, safety, enjoy- ment (from departure to the end of the jour- e. How much of your income do you spend ney). on transport? • Coping strategy with negative factors (e.g., • What decisions do you take about manag- change of schedules, issues if traveling ac- ing money in your home? companied, etc.). • Find out if the interviewee participates in • Have you ever witnessed any instances of such decisions, and how he/she feels about harassment? participation in decision-making. 7. If applicable: After you began your rela- 10. (Based on previous answers) If the inter- tionship/ had children, did you make any viewee work/worked ([find out latest work changes in how and when you commute? experience]: 8. You told me about some difficulties you a. What type of activity(s) did you perform, face when commuting [cite previous exam- and how did you get that job? ples]. Do you think those difficulties may be b. Where is/was that work, and how do/did different for men and women or may affect you commute to it? [find out the mode of women in particular differently? transport, length of trip, cost, difficulties en- countered] PART 3: Experiences, perceptions, c. How is/was your day-to-day experience of work travel when you are/were working? [find out about mobility issues if they haven’t been 9. Now let’s talk about your work experienc- already explored]. es and opinions. I would like to know about your household income. 77 ANNEX d. What was positive, and what was negative your life ten years from now, what type of about working? experience you would have liked to have? [Ask an open-ended question and wait. Clarify e. What is/was your family’s opinion about issues that might not have been mentioned]. your job? [find out the relatives’ opinions: parents, children, partner]. 15. Do you have a plan or strategy for achieving the goals you have mentioned? [explore all f. [if applicable]. What factors led you to the answers] leave your job? a. Do you think it will be possible to achieve 11. If you do not work: For women: Do you ever your dream? [interviewee’s self-discipline go looking for a paid job? and self-esteem at different stages]. a. What are some of the factors that helped b. Can you tell me about any difficulties you you to/impeded you from finding employ- might have to achieve these goals? [Ex- ment? [Discover whether pregnancy influ- plore the range of challenges mentioned] enced job-seeking]. c. What approaches are you using to over- b. How is your life affected by not having come these challenges? your own job and income? 16. Do you think there are any differences be- 12. For men: What do you think about your tween Villa 31/3 de Febrero and downtown partner/daughter/sister/other women Buenos Aires’ job opportunities? working or not working? 17. Think of your perfect job. What would it be a. What is good, and what is bad about your like (where, what, what workload)? partner working/not working? 18. For women: In your opinion, what would b. What is your day-to-day life like? be the reasons that would lead a woman c. Financial consequences and labor divi- to look for a job within Villa 31/3 de Febre- sion at home (for the interviewee and his/ ro, and what reasons would motivate a her family)? woman to look for one outside Villa 31/3 de d. Have the women in your life worked or Febrero? not? 19. Do you admire anyone who is a role model, 13. For men: What do you think would be the an inspiration, for you? Who inspires you, perfect job for your partner/daughter/sis- and in what way? [Ask an open question - if ter? Where, and with whom, would it be? it starts with a famous person, find out who the family and friends admire] [Reasons for preferring specific activities, working away from home, workload?]. • Does that person work, and where? • [Find out about the influential person’s mo- PART 4: Aspirations and Future Plans bility patterns -how does he/she commute, and where to?] Ask the question, “What do 14. Now I want to talk about your thoughts you think of this kind of routine?. and plans for your future. Thinking about the immediate future, also looking back on 78 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES PART 5: FINAL COMMENTS 20. Our conversation was very good, and we were able to talk about many things. Do you have anything else you would like to tell me or ask me? Well, we came to an end. I want to thank you for your time and participation in our study. Your answers will help us a lot for recommend- ing public policies and actions. Thank you very much once again. 79 ANNEX ANNEX 2: DESCRIPTION OF FIELD SITES 1. Buenos Aires pecially concerning building heights, are largely ignored. Two areas of the city of Buenos Aires - Villa 31 and Fuerte Apache - were selected for this During the last military dictatorship, many study, both with low-income profiles but with Villa 31 families were evicted and transferred different accessibility to job opportunities. Villa to Ciudadela in the Tres de Febrero district. The 31 forms part of the Recoleta and Retiro neigh- area was later re-populated by former settlers borhoods, and Fuerte Apache belongs to the and newcomers from the interior of Argentina Tres de Febrero district. as well as from neighboring countries. A plan to regularize tenure for the villa’s inhabitants met only partial success in the 1990s. In 2010 a law Villa 31 and 31 bis (High accessibility) was passed to promote the urbanization of the area (which occupies government-owned land) by the Buenos Aires City Government. The latest estimates put the population at 43,190 (52% female and 48% male). Different demographic dynamics mark the distribution by age and place of birth of the inhabitants of Villa 31 from the rest of the city. Average house- hold numbers are, for example, significantly higher in Villa 31, with 3.27 persons per house- Villa 31 (now Barrio 31) emerged in the mid- hold, versus 2.5 per household in the rest of the 1930s when the government set aside a small city. The average age in the Villa is 23.3 years area to house poorer, mainly Italian immi- (16 years less than the city average), while for grants. It is now known as the “immigrant” women it is 17.8 years. Around 70% of the Villa’s sector. At the same time, an area occupied by population is under 30 years old. railway workers grew up around the “Saldías” In terms of employment and income, there is train station. In common with the city’s oth- a notable difference between men and women. er “villas,” the population of Villa 31 increased Women have much lower incomes in all quin- rapidly during Argentina’s export substitution tiles, except for the lowest, and their formal un- phase, which witnessed massive migration employment rate is 7% higher than men’s and from the country’s interior to the cities. Villa 9% higher in terms of informal employment. 31 consists of 5 neighborhoods: YPF, Comunica- Half of the Villa’s inhabitants originate from ciones, Güemes, Imigrantes, and Autopista. The other countries. 29% were born in Buenos Aires more recent addition to Villa 31, known as Villa (in the rest of the city, more than 60% are native 31 bis, is located between the autopista and the Argentines). Half of the population in the Villa railway tracks. Planning and safety norms, es- 80 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES are Paraguayans, a third Bolivians, and around Avenida General Paz that divides the city from 20% Peruvians. the Province of Buenos Aires. Meanwhile, Ejér- cito de Los Andes occupies about 26 city blocks. Ejército de Los Andes/Fuerte Apache The neighborhood emerged as the result of the Villas Eradication Plan in 1968, aimed at relo- (Low accessibility/ Low Income) cating the inhabitants of Villa 31. The first in- habitants settled in 1973. The journalist José de Zer, covering a shootout in the area, renamed it Fuerte Apache (Fort Apache), a name which soon became a password for insecurity – and a source of stigma, which is evident in the report. Mobility Profile The capacity to generate trips depends on the income group. In Table 1, we can see how the Tres de Febrero is located in the northwest number of trips generated increases in line of the City of Buenos Aires a few blocks from with income in Buenos Aires. TABLE 1. Trip generation rate by income group and district Number of persons Actual trip Income Group Number of trips Number of persons Trip generation rate traveling generation rate Low- 5,044,683 3,709,823 1.36 2,179,454 2.31 Low-medium 4,065,749 2,779,370 1.46 1,721,236 2.36 Medium- 3,675,984 2,408,484 1.53 1,555,783 2.36 Medium-high 3,518,506 2,147,475 1.64 1,460,013 2.41 High income 3,463,060 1,940,732 1.78 1,420,093 2.44 Total 19,767,983 12,985,885 1.52 8,336,579 2.37 Source: Encuesta Domiciliaria de Movilidad de los Hogares 81 ANNEX Table 2 shows the trips made within the City Around 70% of the trips generated in the of Buenos Aires, between the city and the prov- metropolitan region last for 30 minutes or more. ince of Buenos Aires, and those that made with- 52% of commuters on public transport need to out entering or crossing the city. 61% of all trips walk between one and five city blocks to access are made in the Province of Buenos Aires and public transport services. only 24.6% within the city itself. As for private transport, use by low-income TABLE 2. groups is 8% less than its use by people with higher incomes. Travel times are also signifi- Origin and destination of trips cantly shorter for private transport users (av- Origin/Destination Frequency % erage of 60 minutes by public transport, and 20 Within the city 4,861,365 24.6 minutes for over 80% of the people who use pri- vate transport). City-Metropolitan 2,786,224 14.1 Within the metropolitan area 9,311,155 47.1 Gender, travel, and job access Within the district 2,809,239 14.2 In 2014 the World Bank analyzed the 2009 Source: Encuesta Domiciliaria de Movili- Household Mobility Survey for the Buenos Aires dad de los Hogares Metropolitan Region to explore the differenc- Reasons for travel are primarily for work es in travel patterns between men and women (37.4%), followed by study (25.1), which togeth- (Peralta, 2014). er account for more than half of the trips. Regarding modes of transport, a spatial The modal distribution shows that 43% of analysis highlighted a key difference between the trips are done on public transport. Non-mo- the travel patterns of men and women in Bue- torized transport represents 31% of all the nos Aires: they traveled at different speeds. Al- trips. 80% of the population use buses to trav- though average travel times for both sexes are el around the city, and most trips are confined fairly similar, the distance men travel at the to the originating area. Low-income people same time is considerably longer than the dis- use public transport slightly more than high- tances traveled by women. The similarities in er-income groups. Usage decreases in line with average travel times conceal a significant differ- increasing income, a reflection of the impor- ence between trip lengths: 6.72km for men and tance of affordable transportation for the more 4.77km for women. Net travel speeds, estimat- vulnerable groups. Non-motorized transport ed by dividing travel distances by travel time, is standard among those who cannot afford illustrate the difference: 8. 62km/hr for wom- to travel by other means. Poorer people make en and 10.93km/hr for men (26.8% % higher). three times more use of this mode than high- Figure 1 shows that this is especially the case er-income groups. Trips are, on average, shorter for women with children who travel to work. than those taken using different modes (around At 9.7km/hour, the time taken by women is 22 80% last under 20 minutes). % slower than men’s speed (12.2km/hour). The biggest differences in speed of travel are found when comparing working men with women ac- companied by children. 82 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES FIGURE 1 Trip rates, times, distances, and speeds for working and non-working men and women. All Trips. ENMODO 2009. 2.5 Trip Rate 2 1.5 1 0.5 0 50 Average Time 40 (min) 30 20 10 0 10 Average Distance 8 (km) 6 4 2 0 14 Average Speed 12 10 (km/h) 8 6 4 2 0 Without With Without With children children children children Working Non-Working Woman Man Meanwhile, Figure 2 displays the hourly num- the day. This group would be most affected by ber of trips made by men and women through- the limited frequency of public transportation out the day, according to the trip purpose. services at off-peak times. Most off-peak trips Non-working women appear to travel more made by both sexes are not work-related but during the day, particularly in the middle of consist of school runs, household errands, and social visits. 83 ANNEX FIGURE 2. Work-related trips follow a broadly similar pattern: women tend to travel shorter distances Number of trips (in thousands) for in approximately the same amount of time as women and men according to trip purpose. ENMODO 2009. men. This is particularly the case with women accompanied by children who have about the 1.200 same commute times but work in places that 1.000 are 20% closer than those that men travel to. This would suggest that working women with 800 children tend to have jobs within a radius that 600 involves 20 % less than the distances traveled 400 by men. 200 Differences between men and women in - terms of time and distances traveled may re- 0 2 4 6 8 10 12 14 16 18 20 22 strict limit the number of accessible jobs for both groups (estimated at as many as 900,000 W-Work & No Child M-Work & No Child jobs foregone). Figure 3 shows that men with W-Work & Child M-Work & Child W- No Work & No Child M- No Work & No Child children have greater access to jobs than wom- W- No Work & Child M- No Work & Child en with children. FIGURE 3. Percentage of increased accessibility to employment opportunities for men without children compared to women with children within the radius of the AMBA region census area. Source: Logit, background map: OpenStreetMap by CIRED. 84 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 2. Lima Caja de Agua Chaclacayo High accessibility Lima Norte Low accessibility Low income Lima Centro Low income Lima Este Lima Sur Callao Two areas of the city of Lima, Caja de Agua and FIGURE 4. Chaclacayo, were selected due to their low-in- Population distribution, by sex and come profile but with different job accessibility. district (Source: INE) Caja de Agua is a community in the San Juan de Lurigancho district, while Chaclacayo is a dis- 12% trict located in the River Rímac valley. 10% In 2015 the population of the two districts 8% stood at 43,428 in Chaclacayo and 1,091,303 in 6% San Juan de Lurigancho (INE). The population in both districts continues to increase steadily. 4% 2% The population in the two districts is young- er than the average in the Lima Metropolitan 0% 80+ Region.70% of the inhabitants are under 34 0- 4 5- 9 10 - 14 15 - 19 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 65 - 69 70 - 74 75 - 79 years old. Poverty levels present significant dif- ferences: in Chaclacayo, the incidence of pover- ty is 10.1 % (seven points lower than Lima aver- Chaclacayo age), while in San Juan de Lurigancho, it stands San Juan de Lurigancho at 27 % and in Caja de Agua 24.4 %. The latter is Lima one of the poorest districts in the metropolitan region. 85 ANNEX TABLE 3. Employment situation by sex   18-29 30-44 2017 Women Men 45 years + years years Working outside the home 43.7 25.6 63.4 37.2 57.8 36.7 Studying 9.4 8.6 10.2 26.3 1.3 0.2 Studying and working 3.8 3.6 4.0 9.2 2.1 0.0 Working at home 10.1 12.5 7.4 6.0 11.0 13.3 Unemployed 4.3 2.2 6.5 6.0 2.3 4.5 In charge of the household 23.8 45.3 0.4 15.1 25.4 31.0 Retired 4.8 2.0 7.9 0.2 0.0 14.2 NR 0.2 0.2 0.1 0.0 0.2 0.3 Source: Lima Cómo Vamos Around 90% of the population of San Juan de Lurigancho is in the “marginal” (38.7%) or very low-income category (50.1%) (the lowest income levels of the city). TABLE 4. Per capita income in San Juan de Lurigancho (in Peruvian soles) Income group Benchmark in local currency Persons (%) Households (%) High 2,192.00+ 0 0.0 Medium High Between1,330.10and 2,192.20 1.9 0.0 Medium Between 899.00 and 1,330.09 32.8 33.7 Medium Low Between 575.70 and 898.99 44.3 42.5 Low 575.69 or lower 21.0 21.8 Total 100.0 100.0 Source: INE 2013 86 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Map of Caja de Agua This study selected the whole district of Cha- of transportation in the city of Lima and neigh- clacayo rather than a more confined neighbor- boring Callao. Of all the different travel modes, hood, given the fact that it is located in the rela- walking is the most commonly used, which tively fertile valley of the Rimac, and as a result, highlights the critical need to improve public is partly urban and partly rural. Although the spaces, including pedestrian areas, to achieve socioeconomic levels of the inhabitants vary, comprehensive mobility solutions. When peo- the study focused on low-income people living ple were asked about the multimodality of their in small communities scattered throughout the trips, 60.9 % replied that their daily commuting district. consisted mainly of walking, followed by the use of buses or combis (Lima Como Vamos sur- Mobility Profile vey). Women walk, on average, 7% more than men. Collective transport (used for travel to work or educational institutions) is the primary mode 87 ANNEX TABLE 5. Transport modes used in daily trips, by sex and age (in %) Modes of transport Total Women Men 18-29 years 30-44 years 45+ years Walk 60.9 65.1 58.6 64.80 57.9 57.7 Bus 42.3 44.4 41.2 43.2 42.1 41.0 Combi or coaster 37.9 42.6 35.5 39.0 36.1 38.9 Moto-taxi 13.3 13.0 13.5 13.3 13.4 13.0 Car 11.3 6.1 14.0 5.9 16.3 13.8 Colectivo 8.1 9.5 7.3 8.3 8.9 6.3 Metropolitano 6.0 6.1 6.0 6.1 6.1 5.9 Taxi 6.0 7.7 5.0 5.7 6.6 5.4 Metro 4.1 4.2 4.1 3.6 5.8 2.5 Complementary corridors 2.9 3.2 2.8 3.8 2.9 1.3 Motorcycle 2.2 1.1 2.8 2.3 2.6 1.3 Source: Lima Como Vamos 2017 As for average trip times, there are differenc- An average of 23% of the daily trips on the es between men and women. Women spend less city transport network (and more than 25% on time on travel than men and typically tend to feeder routes) are made by people traveling take a lot of very short trips (under 30 minutes) from San Juan de Lurigancho. As an example, or, on average, slightly longer trips of between 30% of passengers on metro/subway Line 1 16 minutes and one hour. The time distribution (Linea 1) are from San Juan de Lurigancho. De- for men is more evenly distributed (Table 7). mand from the district has increased since the line was extended to the far northern end of the TABLE 6. neighborhood. Travel time by sex (%) The map shows Caja de Agua as one of the first stops on the new extension of Line 1. Its Travel times Total Women Men metro station is one of the most crowded in the Less than 15 min 16.30 17.8 15.50 district. 16min-30 min 22.60 21.3 23.3 31min-45min 17.20 20.5 15.5 46min-1h 19.70 20.8 19.1 1h-1.30h 17.10 13.2 19.1 1.30h-2h 5.80 4.9 6.4 2h-3h 1.20 1.6 1.0 Source: Lima Como Vamos 2017 88 FIGURE 6. FIGURE 5. Villa el Salvador Parque Industrial Pumacahua Villa Maria Maria Auxiliadora San Juan Atocongo Jorge Chavez 2012 Ayacuchu Cabitos Angamos San Borja Sur 2013 La Cultura Nicolas Arriola Gamarra Miguel Grau 2014 Source: GYM Ferrovias Map of Lima Metro Line 1 and Street Map of Caja de Agua Demand evolution along Linea 1, by stations (2012-2015) El Angel Presbitero Maestro Caja de Agua Piramides del Sol 2014 Los Jardines Los Postres San Carlos San Martin Santa Rosa Bayovar WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 89 ANNEX FIGURE 7. Annual passenger numbers, by station 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 VES PIN PUM VMA MAU SJU ATO JCH AYA CAB ANG SBS CUL NAR GAM MIG ELA PRE CAA PIR JAR POS SCA SMA SRO BAY 2013 2014 Source: Informe de Desempeño de la Concesión del Sistema Eléctrico de Transporte Masico de Lima y Callao, Linea 1. On the other hand, Chaclacayo is located at around kilometer 27 of the Central Highway (Car- retera Central), one of the country’s most strategic roads for transporting goods and people from the central Andes and eastern areas of Peru to the capital, and the country’s largest port of Callao. The Central Highway has no mass public transport service, although the last station on Metro Line 2, currently under construction, will be located in the neighboring district of Ate. Source: Google Maps 90 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES Gender and Job Accessibility FIGURE 8 Percentage of increased accessibility to employment opportunities for men relative to women at the radio OD zone level in Lima. Data sources: OD survey, background map layer: OpenStreetMap. Differences in travel distances between men35 ever, the 22.0% increase in distance traveled and women36 limit the number of accessible represents a significant difference in terms of jobs37 for each. Figure 8 represents the increased employment opportunities. Altogether, the map employment accessibility of men relative to shows that this difference in average commut- women. The map shows the spatial distribution ing distances translates overall in much more of this unequal access to employment oppor- substantial differences in employment oppor- tunities. In some areas, the shorter distances tunities. don’t imply a massive decrease in employment The spatial pattern of the differences in accessibility either because a large number of the magnitude of the unequal job opportunity jobs are available nearby or because traveling is not apparent. It is, however, easy to see the further does not radically increase the number substantial differences in job opportunities of available formal posts. In other places, how- 35 Average commuting distance of 7.2 km. 36 Average commuting distance of 5.9 km. 37 According to the database, there were 1,435,658 formal employment opportunities. 91 ANNEX FIGURE 9 Absolute increase in accessible employment opportunities for men relative to women at the radio OD zone level in Lima. Data sources: OD survey, background map layer: OpenStreetMap. (>+80%) between men and women in various Figure 9 shows a very different spatial pat- parts of the urban area. This results both from tern. As there are many fewer job opportunities low numbers of jobs available locally and the in the outskirts of Lima, absolute differences in existence of employment opportunities in ad- job opportunities seem to indicate less inequal- jacent radios, which without necessarily giving ity far from the city center. Although Figure 8 access to many more jobs, will change the re- is a better measure for inequality in job access sults in relative terms. In the city center (west between genders because what counts most is side of the city), job accessibilities are quite relative and not absolute accessibility, this map equally distributed (<+20%) as there are large is a useful complement. numbers of jobs available locally. 92 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES 3. Rio de Janeiro Jardim de Fonte Tavares Bastos Low accessibility High accessibility Low income Low income Two areas of the city of Rio de Janeiro were se- creased over time. Pero and Mihessen (2013) an- lected for this study due to their low-income alyzed the proportion of family budgets spent profile and different accessibility to jobs. These on transport spending and the percentage of two areas are Tavares Bastos and Jardim de workers who spend more than an hour on the Fonte. Tavares Bastos is a favela (slum) located way home from work in the RJMR, based on in Morro da Nova Cintra, in the neighborhood data from 2003/200838. For the population with of Catete, in the city of Rio de Janeiro. Jardim low-income or living in peripheral areas, the de Fonte is a neighborhood in the city of Quei- authors found that the region had much worse mados, part of the Rio de Janeiro Metropolitan mobility indicators than any other in Brazil. Region. A study from ITRANS (2003) showed that peo- ple from poorer households in the RJMR make Mobility profile of the MR significantly fewer trips. Two explanations for the negative correlation between the number The state of Rio de Janeiro has one of the worst of trips and income levels are the restricted mobility indicators when compared to other access to motorized transport and the high regions of Brazil, and the differences have in- percentage of income spent on transportation: 38 The authors use the indicator of time spent in commuting from home to work from the 2010 Census, and the burden of urban transport spending on family income, from the Household Budget Surveys (POF) carried out by IBGE (2002/2003 and 2008/2009). 93 ANNEX the poorest 10% of the population in the RJMR the RJMR, based on OD data40. Women’s partici- tie up around 22% of their incomes on trans- pation in the formal job market is substantially port compared to 15.5% of the 20% wealthiest inferior to that of men – while the percentage of (Itrans, 2003; Scovino, 2008). women who do not work is twice that recorded for men. Moreover, people who earn less depend much more on public transport in the RJMR An additional reason for gender dispari- than higher-income individuals. Data from the ty can be found in the different access to and origin/destination survey show that, when con- use of motorized transport modes, as shown in sidering work-related trips, those who receive Table 8. Men are much more likely to use mo- up to two minimum wages represent 70%% of torized transport than women. Furthermore, all travelers using non-motorized and collective again according to the OD survey, women use modes of transportation (PDTU, 2015). other types of transport more than men: e.g., municipal bus, executive bus, pirate bus, min- Low-income women’s mobility in the ivan, school transportation, private car, taxi, moto-taxi, walking, and the metro/subway. RMRJ TABLE 7 There are significant gender differences in mo- Activities of men and women with bility patterns of the metropolitan region where incomes below two minimum wages women face more mobility limitations than according to the OD survey (2012) of men in the same income category. The OD sur- the RJMR. vey (2012) shows that women’s mobility index39 was 9% lower in comparison to men’s (1.36 for Activity Men Women women vs. 1.48 for men). The ITRANS (2003) Formally Employed 39.0 23.2 analysis indicated that in households with up Informally Employed 17.3 14.1 to three minimum wages, women’s mobility in- Unemployed 20.0 41.5 dex was 29% lower compared to men’s (0.8 for Transport Worker 1.3 0.2 women vs. 1.12 for men). Not applicable 19.4 17.5 Women’s lower mobility can be partly ex- Not Registered 3.0 3.5 plained by the lower participation of women Total 100.0 100.0 in the economically active population. While 48 % of men’s travel is work-related (going to Source: Authors’ compilation based on work and/or job-seeking), this figure for women the OD survey (2012). represents only 19% of their trips, with study, health, and grocery shopping, accounting for 38% (ITRANS, 2003). Table 8 shows a compar- ison of work status by gender for individuals with incomes of below two minimum wages in 39 The mobility index is measured by the average number of trips per capita / day. 40 The data includes a household survey that verifies the socioeconomic, demographic and mobility profile of the residents of the selected census tracts, and records all their trips on one day of the week. 94 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES TABLE 8 TABLE 9 Mobility Indices by sex for motorized Mobility Index by sex for motorized and non-motorized trips in RJMR. and non-motorized journeys in three favelas in RJMR Ride Men Women Motorized 1.6 0.91 Ride Men Women Non-motorized 0.42 0.45 Motorized 0.81 0.67 Total 1.48 1.36 Non-motorized 0.90 1.06 Total 1.72 1.73 Source: Authors’ compilation based on the OD survey (2012). Source: Adapted from Koch, Lindau, and Nassi (2013). While no comprehensive survey has been conducted on the specific mobility patterns of favela dwellers, a study conducted by Koch, Tavares Bastos, City of Rio de Janeiro, Lindau, and Nassi (2013) offers some data for RJMR comparison purposes. The authors collected mobility data from three favelas in Rio de Janei- Tavares Bastos is a favela located in the south- ro: Complexo da Penha, Batam, and Babilônia/ ern part of the city of Rio de Janeiro. It is con- Chapéu Mangueira. They found that the total sidered in this study to be an area with high mobility indexes of men and women in the accessibility to jobs. Its foundation dates back favelas studied differed by only 0.01 daily trips/ to 1964. At that time, the favela contained 235 person, in contrast to the data on the metro- families occupying 220 dwellings. In 2010, the politan region as a whole, which showed sig- IBGE demographic census recorded a popula- nificant gender differences. The authors also tion density of 328.03 inhabitants per square ki- found that women in favelas rely more on lometer (versus 5,265.82 inhabitants per square walking than men and that men are more likely kilometer for the city as a whole), with a total to make motorized trips, either in an individual of 1,100 residents occupying 338 dwellings. Of vehicle or using public transport (Table 9). Men these, 51.8 % were recorded as female and 48.2 are also twice as likely to travel using individual % as male. vehicles as women, perhaps suggesting a gen- The gentrification process changed the der disparity in access to “individual” transport structure of the dwellings and the profile of the modes. population. Real estate price increases encour- aged the construction of buildings for rent as well as the “verticalization” of the favela (also due to restrictions placed on further expan- sion of the area). An average building has three floors, with some six floors high. Some of the al- leys and streets are extremely narrow and dark, due to the height of the buildings. 95 ANNEX FIGURE 10 Streets and main services and businesses Source: Google Earth image (13 May 2017). Tavares Bastos has two main thoroughfares: nity consists of informal combis and moto-tax- Rua Tavares Bastos and Rua Cruzeiro do Sul. is. When residents reach the bottom of the hill, Rua Tavares Bastos gives access to the BOPE HQ there is ample public transport available in the and is marked by the red line on the map. The Catete district, including municipal bus lines, other thoroughfare is Cruzeiro do Sul, marked interstate buses, as well as metro/ subway sta- in blue. tions. Table 10 highlights the distance and time required to access the main public transport al- The favela has only one street to access the ternatives. Catete neighborhood: the Rua Tavares Bastos that connects to Rua Bento Lisboa at the foot TABLE 10 of the hill. Within the community, there are Distances and time from Praça Tavares two main “streets” and an alley that is part of Bastos to multiple destinations the Rua Tavares Bastos, which is only suitable for pedestrians. Cruzeiro do Sul Street is wid- Local km Time er (i.e.wide enough for car access from the Rua Rua Bento Lisboa – Catete 1.3 20 min walk Tavares Bastos). It also has a narrower pedestri- Neighborhood Access an-only area). Catete Metro Station 1.7 25 min walk Bus Stop – towards South Zone 1.3 20 min walk Its location at the top of a steep hill, the poor state of the street going up to the favela, plus Bus Stop – toward the Center 1.7 25 min walk the lack of affordable transportation up and Center of Rio de Janeiro 4.8 30 min public transport (at the down the hill, are the leading mobility con- cost of R$3.80) straints that face residents. The residents’ main option to climb and descend from the commu- 96 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES FIGURE 11 Location of the main public transport points Source: Google Earth Image (April 19, 2016). Jardim da Fonte, Queimados, RJMR Queimados residents have a mobility index that is 51% lower compared to the average in- Queimados is a city located approximately 50 dex of the RJMR. A further significant difference km to the northwest of Rio de Janeiro. It is con- between Queimados and the rest of the metro- sidered in this study to be a low accessibility politan area population is the travel time per area in the Rio de Janeiro Metropolitan Region. transport mode. Table 11 indicates that travel Queimados has a population of 137,972, of times for people living in Queimados are signifi- which 52% are female. There are 3.3 people per cantly longer for all motorized modes. household on average. Most of the households have a per capita income below the Brazilian minimum wage41 (R$1050 or approx. US$ 250) (IPEA, 2020). 41 The average household per capita income is distributed as follows: 27% up to half minimum wage; 36% on average earn the minimum wage; 26% between one and two minimum wages; 9% from two to five minimum wages; and only 1% receive more than five minimum wages. 97 ANNEX TABLE 11 General and Motorized Mobility Index General Mobility Motorized Mobility Municipality Travelers Motorized trips Population Index Index Queimados 175,215 71,280 130,872 1.34 0.54 Rio de Janeiro 1,1114,630 8,056,776 5,983,804 1.86 1.35 Metropolitan region 199,156,954 12,529,755 11,279,789 1.77 1.11 Source: Authors’ compilation based on OD (2012). TABLE 12 Modes of transport by district Non-motorized Public transport Private transport Origin Total Trips Trips Trips Queimados 90,090 56 60,282 37 10,998 7 161,369 Rio de Janeiro 3,662,748 31 5,831,329 50 2,225,447 19 11,719,525 Metropolitan Region 7,386,199 37 9,237,844 46 3,291,911 17 19,915,954 Source: Authors’ compilation based on the OD survey (2012). TABLE 13 Average travel times (minutes) Municipality Walking Public transport Private transport Total Queimados 16.5 93.2 44.4 50.0 Rio de Janeiro 17.7 47.5 34.9 35.2 Metropolitan region 17.4 54.8 34.0 37.5 Source: Authors’ compilation based on the OD survey (2012). 98 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES FIGURE 12 Percentage of increased accessibility to employment opportunities for men relative to women at the radio OD zone level in Rio de Janeiro. Data sources: OD survey and RAIS, background map layer: OpenStreetMap. Gender and Job Accessibility42 ployment accessibility either because a large number of jobs are available nearby or because Differences in travel distances between men43 traveling further does not radically increase the and women44 limit the number of accessible number of available formal posts. In other plac- jobs45 for each. Figure 12 represents the in- es, however, the 32.8% increase in distance trav- creased employment accessibility of men rela- eled represents a very large difference in terms tive to women. The map shows the spatial dis- of employment opportunities. Altogether, it can tribution of this unequal access to employment be seen from the map that this difference in opportunities. In some areas, the shorter dis- average commuting distances translates over- tances do not imply a massive decrease in em- 42 A number of caveats need to be mentioned. The maps below use average straight-line commuting distances. All employment opportunities and households are supposed to be located at the centroids of the zone they belong to. Due to unavailability of informal labor database, these analyses do not consider informal jobs. The employment opportunities are considered to be only formal jobs. Some of the distances between centroids are indeed quite large; this can create a threshold effect that would overestimate the inequalities between men and women, especially in zones with large areas. These maps should therefore be interpreted with caution and for illustrative purposes only, not as hard evidence. 43 Average commuting distance of 15.8 km. 44 Average commuting distance of 11.9 km. 45 Data on formal jobs come from RAIS, a national register organized by the Ministry of Labor and Employment. In 2017, there were 2,992,070 formal workers employed in the city of Rio de Janeiro. 99 ANNEX FIGURE 13 Absolute increase in accessible employment opportunities for men relative to women at the radio OD zone level in Rio de Janeiro. Data sources: OD survey and RAIS, background map layer: OpenStreetMap. all in much larger differences in employment ter (east side of the city), job accessibilities are opportunities. quite equally distributed (<+20%) as there are large numbers of jobs available locally. The spatial pattern of the differences in the magnitude of the unequal job opportunity is Figure 13 shows a very different spatial pat- not apparent. It is, however, easy to see that tern. As there are many fewer job opportuni- considerable differences in job opportunities ties in the outskirts of Rio de Janeiro, absolute (>+80%) between men and women exist in var- differences in job opportunities seem to in- ious parts of the urban area. This results both dicate less inequality far from the city center. from the low numbers of jobs available locally Although Figure 12 is a better measure for in- and the existence of employment opportunities equality in job access between genders because in adjacent radios, which without necessarily what counts most is relative and not absolute giving access to many more jobs, will change accessibility, this map (Figure 13) is a useful the results in relative terms. In the city cen- complement. 100 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES ANNEX 3: EXTENDED METHODOLOGY BOX Qualitative Research Methods Qualitative and quantitative research represents different research paradigms. Quantita- tive methods are valuable for drawing conclusions that are valid for the broader popula- tion being studied, and are especially suitable for measuring the frequency of a problem and its distribution in a population. On the other hand, they provide less information about how participants perceive and experience social phenomena and the complex dy- namics involved. Qualitative methods collect data on individual experiences and perceptions. Data collec- tion based on open-ended questions allows the exploration of a range of issues pertinent to a research topic from the standpoint of those who experience social phenomena (in- cluding those that are more subjective or sensitive). However, qualitative methods do not seek to provide an understanding of the distribution of the issues identified in the popula- tion under study. Findings from qualitative research should be understood and used in ways that differ from quantitative findings. The aim is not to test causalities and generalize findings. Findings reveal layers of meaning for a particular group of people, which is important when study- ing human behavior, beliefs, attitudes and perceptions. By studying the conditions, mean- ings and practices found in specific contexts, qualitative research can provide insights that serve to interpret other types of data as well as build and refine theories. Objective. The purpose of this report is to evi- Argentina; and Lima, Peru. In each country, the dence the range of barriers and facilitators that study was implemented in several phases: (1) influence women’s mobility and their access to Study design and stakeholder engagement; (2) economic opportunities in urban contexts. The team training; (3) exploratory data collection ultimate aim is to inform the design of inter- through focus group discussions and participa- ventions that can successfully resolve the lim- tory community mapping; (4) analysis of data iting factors and promote facilitators. from focus group discussions and preparation of research tools for individual interviews; (5) Study phases. The study was conducted data collection through one-on-one interviews in three Latin American metropolitan regions with local men and women and key infor- (MRs): Rio de Janeiro in Brazil; Buenos Aires, 101 ANNEX mants; and (6) data analysis and final write up and one with “high accessibility.” The objective of results. in the sites was to explore the variety of issues that affected individuals with different levels of Sequential Implementation. The research access to work opportunities. Since there was was carried out in the following order of cities: no single tool that could be used in the three Rio de Janeiro, Buenos Aires, Lima. This helped LAC countries to conform to the accessibility cri- to ensure methodological alignment and pro- teria, each country employed different tools to vided an opportunity for teams to submit feed- do this: back on the lessons learned in the course of the study. • The World Bank’s accessibility tool was used in Buenos Aires. This tool factors the use Study design and stakeholder of the public transport system by calculat- ing the prospective travel time for a series engagement of combinations of origins-destinations in Rio de Janeiro team was the first to implement the city. With this data, the number of job the research. The local team reviewed existing opportunities that can be accessed within literature on mobility and gender and second- 60 minutes for each origin-destination was ary data on mobility patterns in the metropoli- defined, and the low and high accessibility tan region. The information was used to develop areas determined. The communities Villa 31 a preliminary research design, which was then and Ejército de Los Andes in Buenos Aires presented to governmental counterparts at the were selected as the high and low accessibil- state level for feedback and validation (Trans- ity neighborhoods, respectively. port Secretariat and Subsecretariat for Policies • In Rio de Janeiro, the IBEU mobility indicator for Women). Next, community leaders were en- was used46. Based on a sample of 290 mu- gaged in each of the two sites selected to provide nicipalities and 1500 neighborhoods in Bra- additional contextual information and support zil, this tool measures the percentage of the the team during data collection. The same pro- population that takes over one hour to get cess was repeated in Buenos Aires where Minis- from home to work in a given location. The tries in charge of the Urban and Social Agenda, communities Tavares Bastos in Catete and Ministerio de Transporte Nacional and Secre- Jardim da Fonte in Queimados were selected atria de Transporte of Buenos Aires city, were as the high and low accessibility neighbor- engaged in the former at the government level, hoods, respectively. According to the IBEU, while in Lima support to reach out to the com- Queimados is the fourth-least accessible munities was provided by local leaders. municipality of 290 municipalities, while Criteria for selection of sites. Data were col- Catete is the seventh most accessible of over lected from six sites according to the following 1500 neighborhoods in the Rio de Janeiro RM. dual criteria: 1) all the sites consisted predom- • Lima: The two target communities were inantly of low-income homes; 2) each metro- selected according to their high or low ac- politan region contained two locations - one cessibility to means of transport, and based with “poor accessibility” to work opportunities on the length of travel times to the places 46 For more details see: ibeu.observatoriodasmetropoles.net 102 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES in the city where most jobs could be found. TABLE 14 The Caja de Agua area in the San Juan de Data Collection (Focus Groups) Lurigancho district was defined as the high accessibility area. In contrast, the districts 18-29 30-50 Total per Type TOTAL years years site of Lurigancho-Chosica and Chaclacayo were defined as low accessibility. Women from low 1 1 2 6 accessibility areas Women from high 1 1 2 6 Team training accessibility areas TOTAL 12 In all the sites, team-training workshops were conducted before fieldwork. These focused on: Themes and methodology. Using a par- a) research design, objectives and guiding ques- ticipatory community mapping methodology, tions; b) research protocols for recruiting and these focus groups aimed to produce a visual obtaining consent from informants, conducting picture of women’s mobility patterns and expe- interviews and ensuring privacy and confiden- riences and identify relevant themes for indi- tiality, team security while in the field, recog- vidual interviews. nizing and responding to situations of violence; c) piloting research tools and organizing feed- • Community mapping. To better understand back sessions. the differences between women’s and men’s mobility and the constraints they face, an Exploratory focus groups and adaptation of the tool of “community map- ping” (WHO, 2005) was developed and used community mobility mapping during the interviews. The exercise starts by asking participants about their daily An FGD guide was developed based on issues routine, and the researchers draw relevant identified in the literature as well as in discus- points and routes on a piece of paper. Then sions with stakeholders. The guide was piloted questions and probing are used to collect in Rio de Janeiro to test the relevance and speci- qualitative data in a visual format on trans- fications of each question, as well as to ensure a port-oriented barriers. “Community mobility coherent sequence of questions. We conducted mappings” include or support the disclosure two exploratory focus group discussions in each of information on information: origins; des- research site (four per country) with women 18- tinations; travel purpose; time-use; routes 29 years old and women 30-50 years old. Each and alternatives; perceptions on the differ- focus group contained 7 to 10 women partici- ent transport modes characteristics (fares, pants from the community and lasted around 1 perceptions of safety, availability, including hour and a half. FGD informants were invited by preference of one modes vis-à-vis other); telephone to participate face-to-face through time use (distribution of activities in the a list obtained from the local government (BA household); urban space. Maps also regis- only), and with the assistance of community tered perceptions of traveling and informa- leaders. A financial incentive was provided for tion on time use, access to employment, those willing to participate in the survey. community infrastructure, social norms, and self-efficacy. 103 ANNEX FIGURE 14 Community mobility map, younger women, low accessibility area, Rio de Janeiro FIGURE 15 Community Mobility Map, high accessibility area, Lima 104 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES TABLE 15 Data Collection (Individual Interviews) Total per metropolitan region Type 18-29 years 30-50 years Total per site TOTAL (HIGH and LOW) Key informants 6 12 36 Women, working far from home 3 3 6 12 36 Women, working close to home 3 3 6 12 36 Women, not working 2 2 4 8 24 Men, working far from home 3 3 6 12 36 Men, working close to home or not 3 3 6 12 36 working TOTAL 204 Individual interviews • Work status: Informants were further sepa- rated into those who worked47 far from home Altogether, 204 semi-structured interviews (i.e., in other cities), those who worked close were conducted with women, men, and key in- to home (in their neighborhoods), and those formants (KIIs). Interview guides were designed who were not working at the time of the based on themes that emerged from the FGDs. study (only applicable to women). The exer- Selection criteria for participants (individu- cise aimed to understand the reasons that in- al interviews). In each of the six sites, the sam- fluenced people to stay in their communities ple was further stratified by: for work reasons or leave the neighborhood in search of better economic opportunities. • Gender: Both men and women were inter- viewed to explore how gender may impact Participant recruitment. The selection of in- mobility and employment decisions; terviewees was based on two strategies: a sys- tematic mapping and snowballing. At the sys- • Age group: Men and women from a younger tematic mapping phase, the interviewing teams (18-29 years) and older (30-50 years) cohort at each site were divided into two groups, each were selected to allow life cycle differences to be located in one of the geographical extrem- to be explored. For ethical reasons, minors ities of the community. Each research group were excluded from the selection. Given that then walked towards the center of the commu- the focus of the research was the relation- nity, making door-to-door calls to identify indi- ship between mobility and access to jobs, viduals who fitted the selection criteria. In the the age of participants was capped at 50 to second phase, snowballing48 was employed to avoid selecting people close to retirement identify working women who could participate, age (and possibly less mobile); 47 “Working” describes any individual who claimed to be engaged in a paid/income-generating (formal or informal) activity at least once a week. 48 “Snowballing” is a non-probability method used when the desired sample characteristic is rare. Informants are selected on the basis of their ability to provide information that may be relevant to the research. 105 ANNEX since most of them were either not at home all ative analysis between countries and groups day or who claimed to be too busy to participate based on a topic’s relative frequency across the in the study. This phase involved approaching different groups. Finally, a consolidated table individuals recommended by community lead- containing data from all the sites was analyzed, ers and people who had already been inter- and the results compared between countries viewed in the same community. and sample groups. Themes and methodology. Topics included: details of informants’ daily routine and mobil- Limitations ity; references to places outside informants’ Certain topics and situations were not explored communities; experiences and attitudes re- in-depth in this study. The sample was limited lated to mobility and work; capacity to make to low-income population groups in urban Peru, appropriate decisions regarding mobility and Brazil, and Argentina (groups living in different work; role models, aspirations, and plans for circumstances from those of wealthier people the future. The interview tool for individual in- or groups in other geographic locations). Safety terviews was also piloted in each site. Interview concerns meant that fieldwork was restricted to and focus group guides were mostly the same certain hours of the day. The sites were classi- in all three countries apart from the language: fied by the United Nations Department of Safe- Portuguese in Brazil and Spanish in Argentina ty and Security (UNDSS) as safe for conducting and Peru. However, field teams in each country fieldwork. Women and men living in areas with were encouraged to probe specific issues during higher levels of crime may face additional bar- the data collection phase. riers to their daily mobilities that are not fully Key informant interviews. These were held captured in this research. Underage individuals with community leaders, transport agency rep- were not included in the samples because pa- resentatives, women’s groups, and academics rental consent would have been needed, thus (the latter supplied information on social, polit- complicating recruitment. Furthermore, given ical, and legal contexts) identified during field- the focus on the relationship between the labor work. market and mobility, we chose not to explore the circumstances of the elderly and retired Data Management and Analysis population. The recruitment strategies may have influ- All the interviews were audio-recorded and enced the profile of informants. The local field transcribed. Transcripts were stored under teams found it difficult to recruit participants unique identifiers in a protected database man- for focus groups and individual interviews aged by the study coordinators. Data was an- since many people would either decline to alyzed in several phases. First, local research participate or fail to show up at the scheduled teams pre-coded data by categorizing relevant times. At the beginning of the fieldwork phase, segments of the transcripts and storing them in the field team did daily rounds in the neighbor- a thematic Excel matrix. Second, all the pre-cod- hoods knocking on the doors of all the houses, ed data was analyzed to produce a consolidated mapping potential participants who fitted the data mapping matrix that summarized topics required criteria, and scheduling interviews. by sampling group, thus enabling a compar- 106 WHY DOES SHE MOVE? A STUDY OF WOMEN’S MOBILITY IN LATIN AMERICAN CITIES The team subsequently used snowballing, and information from community leaders to recruit individuals who were more difficult to recruit, e.g., young men and women who worked and studied (and often faced long commutes) were not willing to be interviewed during their limit- ed free time. Teams encountered challenges due to the study’s multi-country approach. Different local teams in each country collected data. Although steps were taken to reinforce methodological alignment through sequential implementation, kick-off workshops (in each country), regular meetings between local coordinators, and con- tinuous data collection, certain discrepancies in application remained. The research team dealt with these discrepancies during the analysis phase either by not recording findings on top- ics that had not been addressed equally in all three countries or by identifying such results as “country-specific.” 107 ANNEX PARTICIPANT PROFILES SOCIO-ECONOMIC PROFILE BUENOS AIRES LIMA RIO DE JANEIRO Are they single 25% 25% 25% or married? single single single Do they have 77% 25% 25% children? have kids single single How many? 2 2.6 2.3 kids on averages kids on averages kids on averages Do they work? 61% 85% 78% work work work In the district? 18% 50% 50% In the district In the district In the district © 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org