SAFETY FIRST How to leverage social safety nets to prevent Gender Based Violence Operational Guidance Ioana Botea, Aline Coudouel, Alessandra Heinemann and Stephanie Kuttner SAFETY FIRST How to leverage social safety nets to prevent Gender Based Violence Operational Guidance Ioana Botea, Aline Coudouel, Alessandra Heinemann and Stephanie Kuttner ii SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE CONTENTS Acknowledgments v Acronyms vi 1. INTRODUCTION 1 1.1 Rationale 1 1.2 Objectives 2 PART A – Pathways and evidence 4 2. PATHWAYS 5 2.1 Reduced poverty and food insecurity 6 2.2 Women’s empowerment 7 2.3 Increased social capital 9 Risks of SEA/SH and of opportunistic harassment or assault 2.4  9 3. OVERVIEW OF THE EVIDENCE 11 3.1 Cash and in-kind transfers 11 3.2 Public works and economic inclusion programs 14 PART B – Operational guidance 18 4. COUNTRY ENGAGEMENT STRATEGY 20 5. THE ENVIRONMENTAL AND SOCIAL FRAMEWORK 21 5.1 The risk assessment process 21 5.2 Analyzing the prevalence of GBV and key risk factors 22 6. ASSESS 27 6.1 Outreach 27 Intake, registration, and the assessment of needs and conditions 6.2  31 7. ENROLL 36 8. PROVIDE 41 8.1 The transfer of Benefits 41 8.1.1 Payment systems 42 8.1.2 Transfer size, frequency, and duration 49 8.2 Accompanying measures 52 8.3 Provision of services 59 8.4 Public works and economic inclusion programs 61 9. MANAGE AND MONITOR 66 9.1 Codes of conduct 66 9.2 Grievance mechanisms 68 9.3 Monitoring and evaluation 72 BIBLIOGRAPHY 77 ANNEX 1: Additional resources 88 ANNEX 2: SSN project development objectives through a gender lens 92 ANNEX 3: SEA/SH Risk Screening Tool indictors and risk mitigation measures 94  iii FIGURES FIGURE 1. Pathways for SSN impacts on GBV 5 FIGURE 2. Overall effects of cash transfers on IPV (combined results of 22 studies) 11 FIGURE 3. Effect of cash transfers on IPV and controlling behaviors (combined results of 14 studies) 12 FIGURE 4. Social protection delivery chain 19 FIGURE 5. Choosing beneficiaries and recipients of benefits and services 39 FIGURE 6. Payment method design choices 47 FIGURE 7. Design choices: transfer size and frequency 51 BOXES BOX 1. The definitions used in this note 3 BOX 2. 16GBV measurement innovations 16 BOX 3. 17Areas for future research 17 GOOD PRACTICE BOX 4. Designing a program communication strategy 28 EVIDENCE BOX 5. Communicating program objectives and labeling 29 GOOD PRACTICE BOX 6. Designing an outreach strategy in a conservative FCV context 30 EVIDENCE BOX 7. Sex of transfer recipients 38 EVIDENCE BOX 8. Benefit type 42 EVIDENCE BOX 9. Manual vs. digital payment methods 44 GOOD PRACTICE BOX 10. A choice-based payments system in Zambia 46 EVIDENCE BOX 11. The size and frequency of transfers 50 EVIDENCE BOX 12. The duration of transfers 52 EVIDENCE BOX 13. Strengthening women’s social capital 54 EVIDENCE BOX 14. Shifting gender norms 55 EVIDENCE BOX 15. The What Works to Prevent Violence Program 57 GOOD PRACTICE BOX 16. Adapting public works to women’s needs in Ethiopia 63 GOOD PRACTICE BOX 17. Mobile childcare in Burkina Faso 64 GOOD PRACTICE BOX 18. Strengthening South Sudan’s grievance mechanism for GBV mitigation 70 GOOD PRACTICE BOX 19. Establishing a GBV-sensitive grievance mechanism in Zambia 71 BOX 20. Types of monitoring indicators 75 TABLES TABLE 1. Sources for obtaining GBV-related information 23 TABLE 2. Key sources of vulnerability among at-risk groups 24 TABLE 3. Example of a GBV analysis framework 25 TABLE 4. SEA/SH Risk Screening Tool Country Context Indicators (questions 1-13) 94 TABLE 5. SEA/SH Risk Screening Tool project-related indicators(questions 14-25) 96 iv SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE ACKNOWLEDGMENTS This work was financed through a grant from the World Bank’s Rapid Social Response and Adaptive and Dynamic Social Protection Trust Fund Umbrella Program, which is supported by Australia, Denmark, Norway, Sweden, the Russian Federation, the United Kingdom, and the Bill and Melinda Gates Foundation without which this work would not have been possible. The note was drafted by Ioana Botea, Alessandra Heinemann, and Stephanie Kuttner, under the guidance of Aline Coudouel; Claire Cullen and Priyanka Kanth provided research support. Robert Zimmermann edited the report and Jihane El Khoury Roederer designed it. The authors are grateful to peer reviewers for extensive feedback and guidance. At the World Bank, these included Benedicte de la Brière, Margaret Grosh, Mattias Lundberg, Verena Phipps-Ebeler, Changqing Sun, Julia Valliant, and Nahla Zeitoun. External peer reviewers included Amber Peterman and Shalini Roy (IPV Research Collaborative), Emily Esplen and Roopa Hinton (FCDO), Ruth Graham-Goulder (UNICEF), Megan O’Donnell (Center for Global Development), and Tenzin Manell (Women’s Refugee Commission). The authors would also like to recognize the following colleagues who provided detailed technical inputs and feedback on the guidance note: Colin Andrews, Diana Jimena Arango, Loli Arribas-Baños, Laura Campbell, Shubha Chakravarty, Sara Coll-Back, Sara Giannozzi, Rebekka Grun, Emma Wadie Hobson, Junko Onishi, Rohini Pande, Elizaveta Perova, Rachael Pierotti, Aneeka Rahman, Ines Rodriguez Caillava, Manuel Salazar, Nadia Selim, Niyati Shah, Endashaw Tadesse Gossa, Andrea Vermehren, Christopher Walsh, Leora Ward, and Mina Zamand. The authors would also like to thank colleagues who provided feedback on the SEA/SH Risk Screening Guidance Note for Human Development Investment Projects: Yoonyoung Cho, Christabel Dadzie, Matthew Dornan, Ivan Drabek, Jordi Jose Gallego-Ayala, Aylin Isik-Dikmelik, Amjad Zafar Khan, Francesca Lamanna, Matthieu Lefebvre, Muderis Abdulahi Mohammed, Michael Munavu, Edmundo Murrugarra, Junko Onishi, Snjezana Plevko, Aneeka Rahman, Jasmine Rajbhandary, Nina Rosas Raffo, Solene Rougeaux, Marcela Ines Salvador, Maheshwor Shrestha, Julia Smolyar, Changqing Sun, John Van Dyck, Dewen Wang, Asha Williams, Penny Williams, Briana Wilson, and Michele Zini. The guidance note is a component of a larger research initiative in which case studies on social safety nets and gender-based violence were conducted in Bangladesh, Cameroon, Pakistan, and Zambia. The case studies were completed in collaboration with local researchers and with the support of World Bank social protection teams and project implementation staff. In Cameroon, the research team included Stephanie Kuttner, Julienne Ngo Likeng, Kirsten Schuettler, and Thaddée Yossa, with contributions from Irene Jillson. Guidance and support were provided by Francis Batomen, Che Charles, Rebekka Grun, Michelin Njoh, Jacques Christian Pym, and Helène Ndjebet Yaka. In Bangladesh, research was led jointly by Snigdha Ali and Stephanie Kuttner with guidance and support from Rubaba Anwar, Kenichi Nishikawa Chavez, and Aneeka Rahman; data collection and analysis were conducted by the Nielsen Company (Bangladesh) Ltd., under the direction of Farhana Jahan. In Pakistan, the research team included Ahmad Shah Durrani, Stephanie Kuttner, and Mina Zamand, with guidance and support from Amjad Khan, Shahnaz Meraj, and Ali Qureshi. In Zambia, research was led by Stephanie Kuttner and Mpala Nkonkomalimba with guidance and support from Abidemi Coker, Sarah Coll-Black, and Emma Wadie Hobson. Acknowledgments v ACRONYMS ATM automated teller machine CBT community-based targeting ESF environmental and social framework ESSs environmental and social standards FCV fragility, conflict, and violence GBV gender-based violence GDP gross domestic product GEWEL Girls’ Education and Women’s Empowerment and Livelihoods Project (Zambia) IPV intimate partner violence M&E monitoring and evaluation NGO nongovernmental organization PMT proxy means test, proxy-means testing SEA sexual exploitation and abuse SH sexual harassment SIM subscriber identity module or subscriber identification module SSN social safety net vi SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE INTRODUCTION 1.1 RATIONALE Gender-based violence (GBV) has substantial individual and collective costs that disproportionately affect poorer women and girls and can constrain the impact of social programs.1 The numbers of women affected are large and cut across all countries and cultures: one woman in three worldwide, around 736 million, has experienced physical or sexual violence by an intimate partner or sexual violence by a non partner.2 Furthermore, the negative effects of GBV are intergenerational. Children who witness intimate partner violence (IPV)—the most common form of GBV—often suffer lifelong psychological and behavioral problems.3 The cycle then often repeats itself. Girls who witness IPV are more likely to experience violence later in life, and boys who witness IPV are more likely to become perpetrators of violence as adult men.4 In terms of macroeconomic costs, the cost of lost productivity because of domestic violence conservatively ranges from 1.2 percent to 3.7 percent of gross domestic product (GDP)—about the amount most developing countries spend on primary education.5 Globally, GBV is a drain on human capital development, poverty reduction, and growth. GBV also undermines the core objectives of social safety nets (SSNs) by eroding human capital, productivity, and well-being, as well as by increasing women and children’s vulnerability. While reducing GBV is not an objective of most SSNs, these programs are already empowering women and reducing the prevalence of violence against women and children in many cases.6 There is increasing attention on the potential for SSNs to contribute more systematically to the reduction of GBV at scale given their broad reach in many countries. A growing body of evidence suggests that SSNs can lead to an abatement in violence, particularly IPV and violence against children, by reducing poverty-related stress, empowering women, and strengthening their social networks—pathways that are explored in this note. Simple adjustments in the design and delivery of SSNs can amplify the role of SSNs in GBV prevention. GBV is a manifestation of gender inequality and power imbalances between men and women across households and in society generally. Thus, to the extent that SSNs can be designed in ways that empower women and shift gender norms toward greater equality, they can also support reductions of GBV over the longer term. SSN programs can also represent an opportunity for individuals affected by GBV by providing them with resources to leave a violent household and connect them with specialized services. There are also important long-run intergenerational impacts of SSNs that lessen lifetime exposure to GBV risk. Greater educational attainment, smaller age gaps between intimate partners, and delayed age of first marriage are associated with reductions in GBV risks.7 Overall, increasing the human capital of women reduces the risk of lifetime exposure to IPV among the women and their daughters. The COVID-19 1 Women in wealthier households face a 45 percent lower risk of violence (World Bank 2014). 2 WHO (2021). 3 Rivett, Howarth, and Harold (2006). 4 Renner and Slack (2006). 5 World Bank (2014). 6 See annex 2 for an overview of how SSNs with project development objectives that are focused on reducing poverty, increasing social and human capital, responding to shocks, or strengthening resilience provide opportunities to contribute to GBV reduction. 7 World Bank (2014). INTRODUCTION 1 pandemic has reversed poverty reduction and human capital gains and has increased the use of negative coping strategies among poor households. For example, girls are being taken out of school, forced into early marriage, or resorting to transactional sex. SSNs have thus become even more important in global efforts to reduce GBV. It is important to address any risk that SSNs might trigger by exacerbating underlying tensions in households and communities or exposing beneficiaries to risks of sexual exploitation and abuse (SEA). This may be manifested in backlash from intimate partners, SEA by program actors, or opportunistic harassment or assault while traveling to or participating in program activities. In addition to finding opportunities to strengthen positive impacts, it is also essential to identify and reduce any program- related GBV risks, adopt measures to mitigate any residual risks, and monitor the effectiveness of these measures. In fragile and conflict-affected settings and during other emergencies or shocks, such as COVID-19, the risks of GBV are often heightened. Weakened social networks, rising household stress, or escalating violence can increase GBV. In the case of the COVID-19 pandemic, lockdowns have contributed to a global build up in rates of IPV and violence against children.8 Mitigation and monitoring are particularly critical in such situations. 1.2 OBJECTIVES The objective of this note is to provide operational guidance on how to optimize SSN program design and implementation to prevent GBV and empower women.9 It applies to SSN programs that provide regular noncontributory benefits (cash transfers, near-cash, or in-kind transfers), which may include public works or economic inclusion activities. While the guidance focuses primarily on enhancing the direct positive impacts of SSNs, these programs can contribute to GBV reduction indirectly over the longer term by raising educational attainment among girls, reducing early marriage and pregnancy, and other impacts that lessen exposure to violence. Part A provides an overview of the evidence and the pathways through which SSNs influence the risk of GBV. Part B presents operational guidance on the various stages of the delivery chain. Given that the evidence and the practice are emerging, the guidance is not exhaustive and may not apply in all contexts. Rather, it aims to offer a menu of options that may be adapted to the specificities of each program and setting. For the purpose of this note, GBV includes all forms of physical, sexual, emotional, and psychological violence perpetrated by household members or other members of the community (Box 1: The definitions). The note focuses primarily on violence against women because women are the primary targets of GBV. While most evidence of impact focuses on IPV, the operational guidance addresses all forms of violence. In particular, it addresses the SEA of beneficiaries by program actors and sexual harassment (SH) of program actors by coworkers or supervisors. The operational guidance builds on the requirements of the World Bank’s environmental and social framework (ESF) to mitigate project-related risks and, beyond the principle that one should do no harm, to strengthen the positive impacts of programs on GBV prevention.10 8 Reports of violence, including through calls to helplines, have surged by over 25 percent in Argentina, Cyprus, France, and Singapore. 9 This note builds on broad issues identified by the World Bank (2014). 10 Guidelines for conducting SEA/SH risk screening and identifying mitigation measures (World Bank, forthcoming). 2 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE BOX 1. THE DEFINITIONS USED IN THIS NOTE Gender-based violence (GBV) is an umbrella term for any harmful act that is perpetrated against a person’s will and that is based on socially ascribed (that is, gender) differences between males and females. It includes acts that inflict physical, sexual, or mental harm or suffering and threats of such acts, coercion, and other deprivations of liberty. These acts can occur in public or in private. Globally, women and girls are at greater risk of experiencing GBV. However, men and boys may also experience GBV, particularly those who are members of the lesbian, gay, bisexual, transgender, queer, intersex, and asexual community and perceived to transgress ascribed male gender roles. Domestic violence is an umbrella term that refers to all forms of violence within the household. This includes, but is not limited to intimate partner violence (IPV), violence against children, the elderly or persons with disability, and violence by or instigated by co-wives, in-laws, or other family members. Intimate partner violence refers to violence committed by a current or former spouse or partner in an intimate relationship against the other spouse or partner. While IPV can be experienced by men, the majority of IPV is committed against women, particularly the most extreme forms that lead to serious injury and death. IPV is the most common form of domestic violence, although the latter also includes violence against other household members, such as children, the elderly, and persons with disabilities. Sexual exploitation includes any actual or attempted abuse of a position of vulnerability, differential power, or trust for sexual purposes, including, but not limited to profiting monetarily, socially, or politically from the sexual exploitation of another. Sexual assault includes any actual or threatened physical intrusion of a sexual nature whether by force or under unequal or coercive conditions. Sexual harassment includes any unwelcome sexual advance, request for sexual favor, verbal or physical conduct or gesture of a sexual nature that might reasonably be expected or be perceived to cause offence or humiliation if such conduct interferes with work, is made a condition of employment, or creates an intimidating, hostile, or offensive work environment. Opportunistic harassment or assault refers to verbal harassment, threats, or acts of violence against beneficiaries linked to their participation in a program, including while traveling to and from the program site or during program activities, by persons other than intimate partners or program actors. Program actors vary according to the nature of a program and the range of people involved in program implementation. Consistent with the ESF, they typically include (a) direct workers: people employed or engaged directly by the Borrower (including the project implementing agencies) to work specifically in relation to the program; (b) contracted workers: people employed or engaged through third parties (contractors, subcontractors, brokers, agents, or intermediaries) to perform work related to core program functions; (c) primary supply workers: people employed or engaged by the Borrower’s primary suppliers; and (d) community workers: people employed or engaged in providing community labor, such as voluntary services or participation in program activities and processes. Contracted workers (b) include any service providers—individuals, public or private agencies, nongovernmental organizations (NGOs), or firms—hired to implement project activities. They also include private actors, such as employers, firms, or intermediaries, that receive an incentive to hire program beneficiaries or provide them with any other benefit or service. INTRODUCTION 3 PART A Pathways and evidence 4 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 2 PATHWAYS Researchers and social protection practitioners have proposed several direct and indirect pathways through which SSNs can affect the prevalence of violence against women and children. Building on these insights, Figure 1 presents a model of four pathways as the main channels through which SSNs may increase, decrease, or have mixed impacts on multiple forms of GBV. The pathways operate at different levels—within the household, in the community, and during interactions with program actors or participation in program activities. The evidence supporting this framework points to overall reductions in IPV and other forms of GBV. Yet, there are also risks of GBV as a form of backlash, as well as of risks of SEA/SH by program actors and of opportunistic harassment or assault while participating in or travelling to or from program activities. The outcome of any particular intervention is likely to depend on the institutional context, social norms, rates of GBV in the community, women’s bargaining power in the household, the employment status of the intimate partners, and any age and educational gaps between the intimate partners.11 FIGURE 1. Pathways for SSN impacts on GBV Reduced y Reduced poverty-related stress poverty and y Reduced negative coping mechanisms GBV Risks of SEA/SH, harassment and assault food insecurity y Increased emotional well-being y Increased access to and control over resources Direct impacts y Increased bargaining power and status in household Women's empowerment y Increased self-esteem GBV y Risk of backlash to shift in balance of power/ challenges to male authority y Strengthened social networks Increased y Increased status and visibility in the community social capital GBV y Risk of backlash if women are percieved to transgress social norms y Women's increased human capital Impacts Indirect Intergenerational impacts y Increased levels of girls' education GBV y Decreased early marriage and pregancy 11 Baranov et al. (2021). Pathways 5 There are three direct pathways through which SSNs are thought to have impacts on the prevalence of IPV and other forms of GBV: (a) reducing poverty and food insecurity, (b) empowering women, and (c) increasing women’s social capital. There is also a fourth pathway, indirect intergenerational impacts, which is not the focus of this operational guidance. Both the poverty reduction and intergenerational pathways are expected to reduce the risks of IPV in almost all cases. While the women’s empowerment and increased social capital pathways are expected to reduce IPV and other forms of GBV in beneficiary households in the aggregate, there can be mixed impacts in specific contexts or among specific subgroups of beneficiaries, such as between monogamous and polygamous households, or between individuals with varying educational attainment or employment status. Furthermore, in practice, the different pathways operate simultaneously and can reinforce or offset GBV impacts. For example, benefits paid to women may simultaneously challenge men’s role as primary providers, raising the risk of backlash, while also reducing women’s dependence and improving their social status, resulting in an unclear net effect on GBV. Interactions with program actors can both create risks of SEA and provide an entry point for the provision of GBV services. And traveling to or participating in program activities can increase the risk of opportunistic harassment or assault. However, such program-related GBV risks can be reduced, and residual risks can be mitigated through program design and monitoring (see Part B). 2.1 REDUCED POVERTY AND FOOD INSECURITY PATHWAY INTERMEDIATE IMPACTS LEVEL GBV IMPACT Reduced poverty Reduced poverty-related stress and conflicts over Household  and food insecurity scarce resources Reduced negative coping mechanisms Household  (transactional sex, child marriage, alcohol use) Improved emotional well-being Individual  The first pathway through which SSNs are thought to scale back violence within the household is by reducing poverty and food insecurity. Violence within households is often associated with food insecurity, poverty, unemployment, and excessive alcohol use or drug addiction. Violence rates rise as the mental health of household members deteriorates or if men feel unable to fulfill their socially prescribed role as providers, leading some men to express their frustrations through aggression. Men who experience work- and unemployment-related stress are more likely to be depressed, have suicidal thoughts, and use violence against their partners.12 By providing a regular source of income, SSNs are likely to reduce poverty- related stressors and improve emotional well-being, thereby reducing violence within the household.13 Increased access to cash, particularly in extremely poor households, can ease intrahousehold conflict by curtailing arguments over the use of scarce resources and daily spending decisions.14 The predictability and regularity of transfers are important for this effect to become established; stress-related conflicts can resurge if payments are delayed or suspended. 12 Dooley et al. (2019). 13 Ellsberg et al. (2015); Vyas and Watts (2009). Buller et al. (2018). 14 Buller et al. (2018). 6 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Exposure to violence is associated with poverty; however, the relationship is often complex. The association between poverty and domestic violence is bidirectional. Poverty is a key risk factor.15 Moreover, exposure to violence affects well-being and productivity negatively, thereby increasing poverty.16 Women who are poor typically experience greater dependence in relationships with men and less decision- making power in households, which exacerbates the risk of IPV and makes leaving abusive relationships more difficult (see Section 2.2. Women’s Empowerment).17 Poverty also exacerbates other risk factors, including stress and conflict over limited resources, mental and physical health problems, and lower levels of school attendance.18 The sex of the recipient of SSN transfers is not thought to be relevant in this case, because this pathway largely operates through a pure income effect within the household. In addition to lessening violence between spouses, this pathway has the potential to scale back violent or harsh treatment of children. This advance is associated with the improvement in the emotional well-being of caregivers, although cultural norms on disciplining children also come into play. 2.2 WOMEN’S EMPOWERMENT PATHWAY INTERMEDIATE IMPACTS LEVEL GBV IMPACT Women’s empowerment Increased access to and control over resources Individual  (reduced financial dependence) Increased intrahousehold bargaining power and status Household  Increased self-esteem Individual  Risk of male backlash against shifts in the balance of Household  power and perceived threat to male authority The women’s empowerment pathway can reduce IPV if programs transfer resources directly to women or engage women in income-earning activities. SSNs often select women as designated recipients because of their instrumental role in improving human development outcomes. Hence, even if empowerment is not an explicit program objective, transferring resources to women may shift intrahousehold dynamics if the transfers increase women’s bargaining power in the use of resources and in other important household decisions. In focus group discussions among beneficiaries of the Bangladesh Jawtno Program, women reported improved treatment by both husbands and mothers-in-law as a result of their increased status linked to receiving transfers from the program.19 Increased access to own resources reduces women’s dependence and the constant need to ask for money for household or personal expenses, which can lead to conflict, particularly if household resources are scarce. In focus group discussions with beneficiaries of Bangladesh’s Employment Generation Program for the Poorest, women reported that intimate partners were more likely to be cautious about verbal harassment because the women had become 15 Vyas and Watts (2009). 16 Gibbs, Duvvury, and Scriver (2017). 17 Gibbs, Duvvury, and Scriver (2017). 18 The association between violence and wealth is not linear. Instead, the relationship takes the shape of an inverted U in most countries, with a peak in reported violence among the third quintile (Kishor and Johnson 2004). 19 Ali and Kuttner (2020). Pathways 7 income earners.20 Financial independence can also boost women’s self-esteem and status as contributing members of the household, which may reduce their willingness to tolerate violence. Programs that also increase women’s access to financial services may decrease women’s financial dependence, although this outcome may be constrained by discriminatory laws or customs.21 IPV reductions resulting from women’s empowerment may be expected to be sustained to the extent that the empowerment is sustained, that is, if the rebalancing of bargaining power and the reduction of women’s dependence on their intimate partners are sustained. Similarly, less social acceptability of IPV and increased costs to perpetrators would only be expected if the shift in attitudes toward IPV is consolidated. IPV reductions are also more likely to be sustained if improvements in marital relations are long-lasting and lead to enhanced communication and collaboration. However, there are risks of backlash by men to women’s empowerment, particularly among vulnerable groups. Increased bargaining power among women may lead to a reduction in violence because women may become less tolerant of violence if they have a greater ability to support themselves and their children outside of marriage and can exit the relationship.22 However, in many contexts social and legal norms preclude women’s realistic possibility of exiting a marriage even if they have access to sufficient resources, for example, they may risk losing custody of their children. But a shift in the balance of power can also lead to relationship instability and a backlash by men, including the use of violence to prevent women from leaving the relationship.23 If transfers to women elevate women’s status in the household, men may feel threatened and use violence to reassert authority and control. This is especially likely in patriarchal contexts in which women who contribute more to household finances or take jobs defy prevailing social norms.24 In households in which men previously controlled spending decisions or in which different members have distinct priorities, giving cash to women may create competition over the new resources and exacerbate the risk of violence. In particular, if transfers are given to married women, the husbands or other household members could attempt to expropriate the resources, including through threats or the use of violence.25 Transfers to women in polygamous households can also lead to conflicts between wives.26 It is important to balance the risks of male backlash, particularly in conservative contexts, and the risks of reinforcing gender norms associated with IPV. It cannot be assumed, however, that designating men as the primary transfer recipient has a neutral impact on IPV risks. Absent other interventions, by transferring resources to men, SSNs are likely to increase women’s economic dependence on their partners, which is itself a driver of IPV. Thus, while there are risks of male backlash associated with transferring benefits to women, there are also risks associated with strengthening men’s control over household resources if the men are the transfer recipients. 20 Ali and Kuttner (2020). 21 Legal barriers may include, for example, a requirement that husbands must consent before women may open bank accounts. 22 See Farmer and Tiefenthaler (1997); Tauchen, Witte, and Long (1991). Some men may become less violent to avoid that their partners chose to exit the marriage, although it is unclear whether the transfer amounts are sufficient to constitute a credible threat. 23 Eswaran and Malhotra (2011). 24 García-Moreno et al. (2005); Hautzinger (2003); Hughes et al. (2015). 25 Bloch and Rao (2002). 26 Guilbert and Pierotti (2016). 8 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 2.3 INCREASED SOCIAL CAPITAL PATHWAY INTERMEDIATE IMPACTS LEVEL GBV IMPACT Increased social capital Strengthened social networks Individual  Greater status in the community, increasing the social Community  costs to men of perpetrating GBV Shifts in traditional gender norms Community  Risk of backlash against behaviors that transgress Community  social norms A third pathway through which SSNs can prevent GBV is by strengthening women’s social networks and increasing women’s social capital. Accompanying measures, such as behavior change interventions and incentives to access basic services, can contribute to reducing GBV exposure. Interacting with service providers increases the opportunity for women affected by violence to access services. Frequent interactions with other beneficiaries in the community can build women’s social capital. Participating in group-based accompanying measures, such as training or other activities, reinforces social bonds and support networks. These activities tend to boost self-esteem and self-efficacy, especially if life skills modules are included. This can contribute to a reduction in intrahousehold conflict by improving women’s ability to communicate and negotiate effectively for their priorities. Participating in group activities may also make violence more visible and costly for men by increasing the risk of public exposure and social sanctions for misbehavior.27 Improving women’s agency and status within the community may also reduce exposure to GBV outside the household. Program activities that actively involve previously excluded women may strengthen their self-worth, agency, and dignity, thereby improving their ability to bargain, as well as their status more generally in the community. Women who participated in the case study of a cash transfer program in Zambia reported increased feelings of dignity and, thus, confidence in interacting with neighbors by simply having enough money for basic hygiene.28 However, depending on how household and community members react to women’s higher status and enhanced agency, beneficiaries may also face a greater risk of GBV as a form of backlash by members of the community. RISKS OF SEA/SH AND OF OPPORTUNISTIC HARASSMENT OR ASSAULT 2.4  There are cross-cutting risks of SEA in any program in which there are interactions between program actors and beneficiaries. SEA risks can arise if, first, a project actor has influence over decisions affecting a beneficiary and has direct interactions with the beneficiary, but with little or no oversight. As with any abuse of power, risks arise if program actors have decision-making power over a beneficiary (for instance power to decide who is eligible for a benefit) and have the opportunity to use this power during interactions with the beneficiary with little or no oversight (for instance from other beneficiaries, community members, program actors, or local leaders). To reduce such risks, many programs limit the 27 Brody et al. (2015); Pavanello et al. (2016); Stets (1991); Van Wyk et al. (2003). 28 Kuttner and Nkonkomalimba (2020). Pathways 9 actual power of program actors over beneficiaries (for instance, by basing eligibility on objective criteria or centralized systems) or ensure that other individuals are present during key interactions. Any residual program-related risks can then be mitigated through robust grievance systems or other mechanisms able to receive and resolve GBV-related issues. Second, any program activity that requires women to travel, for example, to attend information sessions, seek services, or collect transfers, exposes the women to risks of opportunistic harassment or assault. These can be reduced through careful program design, bringing activities closer to home or creating safe spaces. Third, beneficiaries can also be subject to harassment by other beneficiaries, for instance, while performing public works jointly with men. Fourth, women program actors themselves are at risk of SH by colleagues or supervisors. These risks can be reduced through measures to promote a respectful and safe workplace through codes of conduct and training; residual risks can be mitigated through grievance mechanisms and protections from reprisals against complainants. 10 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 3 OVERVIEW OF THE EVIDENCE Research on SSNs and GBV has mostly focused on the impacts of cash transfers on IPV, with a few studies of impacts on children and adolescent girls. Most of this research does not disentangle the effects of various program components or implementation modalities. While impacts are heterogenous among households with different socioeconomic profiles, the evidence points to positive impacts in the aggregate across multiple forms of GBV and suggests that there are opportunities to enhance these impacts through program design choices (see Part B). 3.1 CASH AND IN-KIND TRANSFERS A growing body of evidence from a cross-section of low- and middle-income countries finds that cash transfers have significant potential to reduce violence against women and children, even if GBV prevention is not an explicit program objective. The results of a review of 22 studies in low- and middle- income countries show reductions in IPV at between 11 percent and 66 percent.29 The mixed-methods review found that (a) 11 of the 14 quantitative studies (79 percent) exhibited declines in IPV attributable to the program; 1 found mixed impacts (both decreases and increases depending on the type of IPV measured); and 2 found no impacts; (b) 5 of the 8 qualitative studies reviewed showed a reduction in IPV after receipt of cash transfers; 1 showed mixed results, with IPV decreasing in some households and increasing in others; and 2 studies showed no clear effect of cash transfers on IPV (Figure 2). FIGURE 2. Overall effects of cash transfers on IPV (combined results of 22 studies) 5 11 2 1 2 1 DECREASE INCREASE MIXED NO RELATIONSHIP IDENTIFIED Quantitative Qualitative Source: Buller et al. 2018. 29 Buller et al. (2018). Overview of the evidence 11 Overall, the impact of cash transfers in reducing the prevalence of IPV is consistent across different forms of IPV. A recent meta-analysis found a significant reduction, ranging from –2 to –4 percentage points, in physical (including sexual) and emotional violence as well as controlling behaviors as a result of cash transfer interventions (Figure 3).30 Of the 14 studies that directly examined the relationship with IPV, none found that cash transfers were associated with a significant overall increase in IPV. Seven of the 14 studies found significant declines in physical/sexual IPV, and the remainder found no significant impact (but with almost all point estimates in the direction of a decrease). The direction of the effects in most of the studies that examined emotional IPV also suggested a decrease; 2 of the 10 studies produced statistically significant estimates. Four studies reported results for controlling behaviors, for instance, the husband restricting the woman’s contact with her family, insistence on knowing where she is at all times, and becoming angry if she speaks with another man; 3 of the 4 found that cash transfers were accompanied by a significant decline in this type of behavior.31 More recent research has found similar results.32 FIGURE 3. Effect of cash transfers on IPV and controlling behaviors (combined results of 14 studies) 8 8 7 1 2 3 PHYSICAL/SEXUAL IPV EMOTIONAL IPV CONTROLLING BEHAVIORS Decrease No impact detected Source: Baranov et al. 2021. These encouraging average effects of SSNs on IPV may nonetheless mask a rise in some forms of IPV among certain subgroups. Women with limited schooling emerge as particularly vulnerable. Evidence from Latin America shows that, although average IPV prevalence among beneficiaries declined after cash transfers were delivered, the prevalence of some forms of violence increased among less well educated women and that the effect was mediated by the educational attainment of their husbands. For instance, although a study in Ecuador found a decrease in controlling behavior, on average, in See Baranov et al. (2021). There is substantial overlap in the studies covered by Buller et al. (2018) and Baranov et al. (2021). In addition to the 30  9 studies covered in both, Buller et al. (2018) include quasi-experimental and qualitative studies, while Baranov et al. (2021) include studies that have been published since Buller et al. (2018). 31 The 14 studies reviewed by Buller et al. (2018) examine 56 IPV indicators, including 34 measures of physical or sexual violence. Across all 56 outcomes, 20 (36 percent) are statistically significant and negative. The remaining 63 percent show no significant change in IPV because of cash transfers. For significant reductions in IPV, the percentage varies by category of violence examined: 44 percent of indicators of physical/ sexual IPV and 38 percent of other outcome indicators (such as controlling behaviors) demonstrate a significant reduction in violence, whereas only 8 percent of emotional IPV indicators do so. The one case where an increase is found in emotional IPV is in the Give Directly pilot initiative in Western Kenya in a comparison of treatment to nontreatment households in the same villages (Haushofer and Shapiro 2016). Furthermore, nine of these impacts represent reductions of 30 percent or more, which is quite notable given that most evaluations took place over the short or medium term.  eductions: Heath, Hidrobo, and Roy (2020); Lees et al. (2021); Roy et al. (2019); WFP (2019) (qualitative in Bangladesh, the Arab Republic of Egypt, 32 R El Salvador, Jordan, and Mali). No impact detected: Haushofer, Mudida, and Shapiro (2019); Haushofer and Shapiro (2018); Litwin et al. (2019). 12 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE beneficiary households, women with less than six years of schooling whose husbands had even lower levels of education experienced a substantial rise in emotional violence.33 In Mexico, if husbands had low educational attainment or no education and were drinkers, aggressive behavior after drinking increased by 30 percent; it increased even more if the wives were younger.34 These studies therefore suggest that there is a greater risk of backlash to women’s improved access to cash or economic opportunities through SSNs if men are already in a weak position. Such violence may stem from the sense of powerlessness or insecurity of the perpetrators, especially if they feel unable to meet the roles socially assigned to them.35 Effects also differ based on household structure, such as variations between polygamous and monogamous households.36 For instance, Mali’s national unconditional cash transfer program had no systematic effects on IPV in monogamous households, but large, significant reductions among polygamous households. In particular, violence decreased against second and later wives; these wives had faced the highest rates of violence in the absence of the program. Conversely, in Ghana, IPV reductions were only observed among monogamous households, while there were no reductions across any domain of IPV in polygamous households.37 The conflicting findings of these two studies may reflect the different sex of the designated recipients—men in Mali and women in Ghana, although more research is needed to establish patterns. In the case of Zambia’s Social Cash Transfer Program (providing support to woman-headed households, a large number of whose heads were older widowed women), qualitative research revealed that the risks of GBV arose mainly from adult sons who used threats of violence to confiscate beneficiary cards or cash transfers. If vulnerabilities related to age and sex intersect, different risk reduction mechanisms may be needed, including direction program communication toward the adult children of beneficiaries.38 Most conditional and unconditional cash transfers studied made the transfers to women. Because most programs only have one modality (and do not include separate modalities for different design elements), it is difficult to isolate the impacts of different design features. However, given that all programs reduced poverty among beneficiary households, it is possible to conclude that a common transmission mechanism through which cash transfers reduce IPV is the lowering of poverty-related stress. Increased economic security and emotional well-being diminish conflicts over scarce resources.39 The fact that some women receiving transfers experience a rise in emotional violence may indicate that husbands confiscate resources or lash out in response to women’s economic empowerment. An early study of the Prospera Program in Mexico found no evidence that husbands arbitrarily took possession of the transfers.40 Subsequent research revealed, however, that a subsample of beneficiaries suffered instead from increased emotional abuse and threats of physical violence.41 In Uganda, a microenterprise training program, coupled with a one-time cash grant, showed a similar increase in controlling behaviors as a result of greater efforts by intimate partners to capture and control earnings.42 These findings suggest that some men may use the emotional violence and threats of physical violence to pressure women into handing over all or a portion of the cash transfers and thereby reclaim their control over household finances. 33 Hidrobo and Fernald (2013). 34 Angelucci (2008). 35 Jewkes (2002). 36 Guilbert and Pierotti (2016) analyze polygamous households in Burkina Faso to see how variations in household structures, intrahousehold dynamics, division of responsibility, and resource allocation may mediate outcomes. 37 Peterman, Valli, and Palermo (2021). 38 Kuttner and Nkonkomalimba (2020). 39 Haushofer and Fehr (2014); Rojas (2011). 40 Angelucci (2008). 41 Bobonis, González-Brenes, and Castro (2013). 42 Green et al. (2015). Overview of the evidence 13 Evidence suggests that SSNs are most effective in reducing the risk of IPV if transfers are accompanied by complementary measures (cash plus). In Bangladesh, decreases in IPV were sustained six months after cash or food transfers ended only if the transfers were combined with accompanying measures on nutrition.43 The sustained impacts were plausibly linked to the increased social capital of beneficiaries resulting from the interactions during nutrition training sessions that allowed the beneficiaries to develop close relationships with each other. A follow-up study found that the reduction in physical IPV from this cash plus intervention persisted four years after the program ended.44 Similarly, in northern Nigeria, while cash transfers alone caused a rise in sexual IPV, the transfers contributed to an overall reduction in IPV if they were combined with a whole-of-village livelihoods program.45 A likely explanation is that husbands and other community members also benefited from improved income and consumption as a result of the livelihoods component, making the transfers to women appear less threatening to their partners. There is also some evidence that SSNs contribute to the reduction of violence against children.46 Drivers of violence against women and against children overlap, and transmission mechanisms for the reduction of the violence can therefore be expected to overlap.47 A systematic review of 14 studies in low- and middle-income countries found that approximately one indicator in five showed statistically significant reductions in violence against children.48 The most promising evidence was related to sexual violence experienced by female adolescents in Africa, while there was less clear evidence of impacts in other regions and on young child measures, including harsh discipline. Similarly to IPV, reductions in violence against children are thought to be mainly the result of (a) increased economic security, leading to less need for negative coping mechanisms, such as transactional sex, and the substance abuse often linked with violence; (b) access to education; (c) reduced poverty-related stress, leading to enhanced psychosocial well-being and caregiving; and (d) reduced intrahousehold conflict. More recent evidence from Tanzania bolsters these findings by showing that participation in a cash plus intervention among adolescents reduces sexual violence against girls and physical violence perpetrated by boys.49 3.2 PUBLIC WORKS AND ECONOMIC INCLUSION PROGRAMS While the payment of wages through public works programs is expected to affect gender-based violence in ways similar to cash transfers, the work requirement can also have impacts on GBV. Public works schemes transfer cash (wages), conditional on the provision of labor. In Bangladesh, public works programs have been found to be more effective than direct transfer programs in empowering women, precisely because of the work requirement.50 The improved household decision-making power identified among women participants has been attributed to the sense of pride instilled among the women in their contribution and ownership of the income earned. Husbands respected their wives more if the wives became income earners, while there had been little appreciation of women’s domestic work. Based on evidence from India, depositing wages directly in woman-owned bank accounts can strengthen the empowerment effect of public works, particularly among women without prior experience of working and women whose husbands disapproved of their working.51 43 Roy et al. (2019). 44 Roy et al. (2019). 45 Cullen, Gonzales Martinez, and Papineni (2020). 46 While there is limited evidence of impacts on other household members, SSNs are expected to reduce the neglect and abuse among these members, including children, by reducing poverty-related stress, enhancing psychological well-being, and improving caregiving. 47 Fulu et al. (2017). 48 Peterman et al. (2017). 49 Based on analysis presented by Lusajo Kajula, Tia M. Palermo, and others at the virtual Cash Transfer and Intimate Partner Violence Research Collaborative–Intimate Partner Violence Initiative workshop on October 29, 2020. 50 Ahmed et al. (2009). 51 Field at el. (2019). 14 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE The empirical evidence of the impact of public works on IPV is mixed. In India, an increase in female labor participation as a result of the National Rural Employment Guarantee Scheme was initially associated with a weakly significant increase in domestic violence.52 More recent analysis found, however, that participation in the scheme mediated the adverse effect of droughts on domestic violence by reducing poverty-related stress within the household.53 Similarly, suggestive evidence from Sierra Leone, where rates of female labor participation are much higher, found that physical IPV declined as a result of the public works program.54 However, in Laos, while participation in public works was linked with improved empowerment, there was no impact on IPV.55 Context emerges as a key heterogeneity factor, but evidence on cash transfers suggests that the relationship between women’s status and IPV is often mediated by the husband’s employment status. Indeed, women’s employment has been found to be associated with greater vulnerability to physical violence if the women have better employment (that is, in regular employment rather than farm, unwaged, or seasonal work) than their husbands, regardless of overall income, educational attainment, or rural or urban residence.56 More research on the impact of public works on IPV is needed. Similarly, the evidence on the impact of community backlash against women engaged in public works varies substantially across contexts. In India, in addition to initially exacerbating IPV, the National Rural Employment Guarantee Scheme led to an increase in kidnappings and harassment of women participants because of greater exposure to risks during the commute and at the unsecured workplaces.57 In Bangladesh, some women participants mentioned that they had been the victims of verbal attacks by other villagers because of their participation in the programs; it was not considered appropriate for women to engage in manual labor.58 In interpreting these findings, it is important to take into account the conservative cultural context of South Asia, where women’s labor force participation and mobility are more highly restricted. For example, qualitative research in Cameroon reveals that women who participate in public works as part of the Projet Filets Sociaux (Social Safety Nets Project) were seen by other members of the community as hard workers who contributed to the improvement of the community’s well-being, thereby raising the women’s social status.59 These different findings highlight the importance of context and prevailing social norms in mediating the relationship between women’s employment and GBV. Empirical evidence on the impact of economic inclusion programs on GBV remains limited, despite a growing body of research on the effectiveness of the programs in improving employment, earnings, and other well-being outcomes among women.60 The evidence that points to no effect on IPV includes the following. In Afghanistan, a multifaceted program was found neither to increase nor to decrease IPV among women participants.61 In Burkina Faso, a comprehensive livelihoods intervention was found to have a nonsignificant reduction in physical IPV.62 An asset transfer and microfinance intervention in Uganda similarly failed to detect impacts on IPV.63 Promising evidence has emerged around economic inclusion programs overlaid with accompanying measures directly tackling gender norms and underlying power dynamics. Nonetheless, a livelihoods training program in South Africa that also included activities addressing gender failed to achieve significant change in the experience of IPV by women, despite a decrease in the reported perpetration of IPV by men participants.64 52 Amaral, Bandyopadhyay, and Sensarma (2015). 53 Sarma (2020). 54 Additional analysis based on data collected by Rosas and Sabarwal (2016). 55 Dervisevic, Perova, and Sahay (2021). 56 Agarwal and Panda (2007). 57 Amaral, Bandyopadhyay, and Sensarma (2015). 58 Ahmed et al. (2009). 59 Kuttner, Ngo Likeng, Schuettler, and Yossa (2020). Economic inclusion programs are multidimensional interventions that support individuals, households, and communities so these can raise 60  their incomes and build their assets. They are also referred to as productive inclusion or graduation programs. See Banerjee et al. (2015). 61 Corboz et al. (2019). 62 Ismayilova et al. (2018). 63 Green et al. (2015). 64 Gibbs et al. (2017). Overview of the evidence 15 Overall, these are important areas for future research, though research on GBV raises important ethical and methodological challenges. Any measurement of GBV calls for precautions above and beyond routine data collection to guarantee that no harm is caused (Box 2). And researchers have developed innovative instruments and approaches for collecting data accurately and safely. In terms of gaps in our knowledge, the previous paragraphs have highlighted many areas for future research, spanning from a better understanding of particular design choices to the impact of accompanying measures, the cost of interventions, and the heterogeneity and sustainability of impacts, in addition to a host of technical issues (Box 3). BOX 2. GBV MEASUREMENT INNOVATIONS There are several ethical considerations in conducting research on GBV, including the following: confidentiality and safety, the need to ensure that the research does not cause any participant further harm (such as trauma), the importance of ensuring that the participant is informed of available sources of help, the need for the interviewers to respect the decisions and choices of interviewees, and the need to minimize distress to researchers. Primary data collection on GBV requires that a mechanism be established to provide referrals for care and support for survivors identified during the interviews, including psychosocial, legal, health care, and police services. Given the unique challenges and sensitivities involved in measuring GBV, including IPV, researchers have developed innovative instruments and approaches for collecting data accurately and safely. Accompanying research has revealed that the reported prevalence of GBV may vary substantially depending on the method of measurement. „ Audio computer-assisted self-interviewing. This is a method of data collection whereby respondents listen to prerecorded questions through headphones and respond by selecting among possible answers on a touchscreen or keypad (for example, a corresponding colored square). Asking questions through audio and handheld tablet interfaces has the potential to improve the measurement of GBV by allowing greater privacy and anonymity. Evidence from the Dominican Republic, Liberia, and Malawi confirms this hypothesis by showing that the method results in more disclosures relative to the direct response method and also increases the reporting on GBV. However, no difference in reporting between this method and face-to-face surveys was detected in the Philippines, which suggests that privacy standards, computer literacy, and other features of the context are important mediating factors.a „ List experiments. This approach assures anonymity by not requiring direct answers to sensitive questions. Instead, respondents are provided with a set of statements and asked to indicate the number of statements they consider true. Half the respondents are randomly assigned to receive only nonsensitive statements, while the other half receive the same nonsensitive statements, plus one sensitive statement. GBV prevalence is measured by the difference in the results between the two groups. New evidence from Nigeria and Rwanda and indicates that the list method produces the highest prevalence estimates (39 percent and 100 percent greater, respectively, compared with direct methods).b However, further research is needed to replicate the analysis and shed light on the circumstances in which list experiments are successful in increasing GBV reporting. List experiments are infrequently used because of the additional implementation challenges. For instance, enumerators required in-depth training and practice to be able to explain the exercise clearly, while respondents may not understand that the method grants them anonymity or they count incorrectly. a. Dervisevic, Perova, and Sahay (2021). b. Cullen (2020). 16 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE BOX 3. AREAS FOR FUTURE RESEARCH There has been a surge in research on the impact of SSNs on IPV over the last few years, including by involving public health experts. While the evidence largely points to positive impacts and isolated subgroup adverse impacts, it is also starting to highlight concrete design features that help mitigate IPV risk and contribute to GBV prevention more broadly through cash transfers. Despite the encouraging progress, several key knowledge gaps remain, as follows:a „ Impact of program design choices: Few studies isolate the impacts of different design features or program components to determine, for example, whether transfer amount or frequency (such as lump sum vs. smaller, more frequent transfers) have diverse impacts on IPV; to evaluate the trade-offs between transferring cash to men vs. to women in different cultural contexts; to determine whether in-person or digital payments are most likely to remain within the recipient’s control; to evaluate the relative impacts of different messaging approaches (such as framing programs around women’s empowerment vs. family well-being). „ Role of accompanying measures: many questions remain about the impacts of accompanying measures. For example, what are the most efficient and cost-effective plus interventions that could be implemented alongside cash transfers to prevent GBV? Given what is known about the positive impacts of non-GBV–focused nutrition sessions in Bangladesh, how do the impacts of similar interventions focused on general human development compare with interventions focused on changing gender dynamics and reducing GBV? What are the specific program elements responsible for the positive impacts of interventions explicitly aiming at norm change and GBV prevention (for example, couples training, edutainment, interventions targeting men and boys, women’s empowerment interventions, women’s livelihoods and self-efficacy interventions)? Are they scalable? „ The cost-effectiveness of proven interventions: SSNs have the potential to contribute to GBV reduction at scale given their coverage as national programs in most countries. However, little is known about what implementing effective accompanying measures at scale would entail. Costing or cost-effectiveness studies should accompany impact evaluations to determine the relative impacts of programs of different GBV-focused and non-GBV–focused accompanying measures compared with cash only transfers. „ The heterogeneity of impacts: How do impacts on GBV vary based on sociodemographic or other characteristics? More research on heterogeneous effects (by vulnerability, family structure, educational attainment of beneficiaries and their partners, and so on), particularly from other regions besides Latin America, would help identify risk factors and inform mitigation solutions. „ Diversity in GBV typologies: How do SSN programs influence violence other than IPV in beneficiary households, such as violence against children or the elderly, and of the GBV experienced outside the household while engaged in program-related activities? „ Long-term impacts: Most studies assess GBV impacts during program participation or shortly after a program ends, but what are the longer-term effects? Alternatively, are there intergenerational impacts of SSNs? „ Measurement: Which survey methodologies generate the most accurate GBV prevalence data and which factors influence the relative accuracy? How might one control for a potential increase in GBV reporting as a result of the intervention that boosts the ability of participants to recognize forms of GBV that have become normalized? What are the ethical considerations in weighing research and measurement methods? Under which circumstances are pure control arms justified from a learning perspective, given the strong evidence on the impact of cash transfers? „ Context and external validity: While context is an important confounding factor, few papers seek to unpack the contribution of context (for example, gender norms) to the relationship between GBV and participation in SSNs. More research is needed to clarify the extent to which current evidence is generalizable to different cultural contexts. a. T  he gaps were discussed at length during a virtual Cash Transfer and Intimate Partner Violence Research Collaborative–Intimate Partner Violence Initiative workshop on October 29, 2020. See Cash Transfer and Intimate Partner Violence Research Collaborative (dashboard), International Food Policy Research Institute, Washington, DC, https://www.ifpri.org/project/cash-transfer-and-intimate-partner-violence-research-collaborative Intimate Partner Violence Initiative (dash- board), Innovations for Poverty Action, New Haven, CT, https://www.poverty-action.org/program-area/health/intimate-partner-violence-initiative. Overview of the evidence 17 PART B Operational guidance 18 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE This section breaks down key choices in program design, the evidence on the impact of these choices on GBV, tips on design and implementation, and examples of good practice. The tips on design and implementation include guidance to address (a) project-related SEA/SH risks as required under the World Bank’s environmental and social assessment (indicated with red bullet points); (b) other forms of program- related GBV, such as backlash within households and communities or opportunistic harassment and assault while participating in program activities; (c) guidance to increase the potential for SSNs to empower women and prevent GBV through smart design choices and operational tweaks; and (d) opportunities to shift norms and address the fundamental drivers of GBV that may require more significant resources and commitment. For many design and implementation features, the best choice will depend on contextual factors; there is no one size fits all. These factors include institutional capacity, counterpart engagement, and the available resources, and other factors that need to be assessed during project preparation. Additionally, there are often trade-offs between the benefits and risks associated with each design choice. For some beneficiaries, there may be a greater GBV risk associated with a specific design element, for example, designating women as the transfer recipients rather than men. However, designating women as the transfer recipients can contribute to women’s empowerment, which may prevent GBV in the longer term and may therefore be the better choice. For any particular design element, there may also be trade-offs between gender- and GBV-related objectives and other program objectives, for example, in determining the size of the transfer. In many other cases, taking gender and GBV into account is simply consistent with good program design generally, such as implementing a well-designed and culturally sensitive communication strategy. More generally, most design and implementation choices that advance the objectives of empowering women and reducing GBV enhance the project’s development impacts by building human capital and reducing poverty among vulnerable populations. To facilitate the practical application of this guidance by World Bank project teams and counterparts, it is structured along the social protection delivery chain (Figure 4). Well-designed SSN programs optimize GBV reduction, while mitigating any program-related risks at each stage of the delivery chain. In addition, two sections of the guidance cover the broader country engagement strategy and the application of the ESF. Each section addresses a phase of the delivery chain and starts by providing a description of the stage and highlighting the design and implementation aspects that may be leveraged to empower women and prevent GBV. Each section then suggests a set of key questions to be considered during the design and implementation of a program. This is followed by an in-depth discussion of the choices along the chain supported by evidence and practice. Each section closes by offering tips on the design and delivery of the program. Boxes presenting evidence and best practice appear throughout, as do decision trees in relation to key design features. FIGURE 4. Social protection delivery chain ASSESS ENROLL PROVIDE MANAGE Eligibility Determination Provision of Bene ciaries Exit decisions, Assessment and of bene ts Noti cation bene ts compliance, noti cations, Intake and of needs and enrollment and service and and/or updating, and and case Outreach registration conditions decisions package onboarding services grievances outcomes RECURRING 1 2 3 4 5 6 7 8 9 CYCLE Source: Lindert et al. 2020. PERIODIC REASSESSMENT PART B | OPERATIONAL GUIDANCE 19 4 COUNTRY ENGAGEMENT STRATEGY GBV is a complex, multidimensional problem that is most effectively tackled through a coordinated multisectoral response. SSNs can provide an entry point for engaging counterparts in a dialogue on a national strategy for GBV prevention and the provision of support services to GBV survivors. SSN programs can also directly contribute to GBV prevention as part of a portfolio-wide strategy and in coordination with other World Bank activities, particularly in human development sectors. The contribution of any single SSN program to GBV prevention should therefore be situated within the broader countrywide portfolio and dialogue. This includes support for policy and program development and for capacity building in national GBV-support services. The country engagement also represents an opportunity to strengthen counterpart engagement on gender and GBV issues generally, particularly in contexts where this is weak. Thus, SSNs can be an important entry point for dialogue with counterparts on gender and GBV issues, in addition to contributing directly to women’s empowerment and GBV prevention. The following operational guidance has been developed from this perspective. In most countries, there is often the need to strengthen the capacity of SSN implementing agencies on gender and GBV and on their intersection with SSN programs. Institutional assessments completed during the early phases of implementation should include assessments of the capacity of implementing agencies to understand gender issues and monitor program-related GBV risks. Capacity-building strategies can then be developed. 20 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 5 THE ENVIRONMENTAL AND SOCIAL FRAMEWORK 5.1 THE RISK ASSESSMENT PROCESS The World Bank’s ESF includes both the environmental and social policy for investment project financing (requirements that apply to the Bank) and the environmental and social standards (ESSs, requirements that apply to the Borrower and to projects).65 The ESF provides the foundation for the improved social risk management through which key social risks and associated mitigation measures are identified. Beyond do no harm due diligence requirements (including assessing, mitigating, and monitoring project-related SEA/SH risks), the ESF offers a progressive framework to maximize project development impacts, including women’s empowerment and GBV reduction. As a critical social risk that may arise in World Bank–supported operations, the identification and management of GBV risks align with a number of key ESSs, including ESS 1: the assessment and management of environmental and social risks and impacts; ESS 2: labor and working conditions; ESS 4: community health and safety; and ESS 10: stakeholder engagement and information disclosure. During project preparation, design and development of World Bank-supported operations should include identification and assessment of key risks that may contribute to SEA/SH and other forms of GBV. This risk assessment process provides an important opportunity to understand a project’s context, identify key drivers of GBV that might interact with SSN programs, and monitor risks and mitigation measures. The risk assessment process, to be conducted by both Bank teams and government partners, provides an opportunity to aggregate available data on GBV in communities and households, including violence committed against spouses, children, the elderly, and persons with disabilities and violence committed or instigated by nonintimate partners, such as in-laws and co-wives. Importantly, it is not advisable to collect new data on GBV prevalence during project preparation because this requires adherence to protocols and standards specific to the collection of GBV information to avoid creating risks among respondents.66 The risk assessment process also offers a mechanism to analyze gender norms, gender dynamics, the drivers of various forms of GBV, and the socioeconomic and institutional framework in which the project will be implemented. This may cover market forces, institutions, laws, policies, and other programs, and whether the program is being implemented in a context of fragility, conflict, and violence (FCV).67 These details can help identify the risks of backlash by a partner who is a man against a beneficiary who is a woman perceived to have challenged the partner’s role as family provider, the risks of assault of women beneficiaries traveling to or from program activities, or the harassment of women at public works sites for perceived transgressions of social norms that fix the acceptable forms of women’s work. It is critical that this analysis take place early in the process to allow it to inform the project design. As part of the risk assessment process, the team is also required to identify mitigation measures proportional to the GBV risks that have been identified, measures that will be monitored as part of the environmental and social 65 World Bank (2017). 66 Ellsberg and Heise (2005). 67 Buller et al. (2018); Peterman, Valli, and Palermo (2021). The environmental and social framework 21 management framework or the environmental and social management plan. The process needs to be continuous throughout project implementation as a mechanism for systematically monitoring both the risks and the implementation of mitigation measures. As part of the risk assessment process, the World Bank has developed guidance that focuses more specifically on project-related SEA/SH risks. This SEA/SH risk identification process is conducted by both the World Bank team and government partners. For World Bank teams, the risk assessment includes applying a dedicated risk screening tool as well as consideration of wider risks, and identifying mitigation measures appropriate for the identified level of risk. The SEA/SH risk screening exercise centers primarily on identifying the potential for SEA by program actors against beneficiaries or for SH by program staff in the performance of their duties, though teams are encouraged to also consider other risks which may have a bearing on these (reflected in the government’s environmental and social assessment). The resulting SEA/SH risk rating contributes to a project’s overall environmental and social risk classification. Mitigation measures proportional to the level of risk are subsequently identified and incorporated into the project design and relevant safeguard instruments, including environmental and social commitment plans, environmental and social management plans, and the environmental and social management framework.68 Annex 3 presents in detail the questions used to assess program-specific SEA/SH risks in social protection programs and the associated mitigation measures and provides links to relevant documents. The risk assessment process further includes consultations with stakeholders to help in the identification of risks and locally relevant measures to address those risks during preparation, and to monitor GBV risks as part of the project’s stakeholder engagement plan. Local women’s groups, groups that advocate for children and adolescent rights, women leaders, and other stakeholders can help clarify the local gender and GBV dynamics. If the ability of women and girls to express their needs and concerns is somehow limited, effective consultation would require that women be given the opportunity to participate separately or in women-only groups. Such discussions should be led by experienced facilitators who are sensitive to local gender and GBV dynamics and should be conducted in line with relevant guidance on the ethical performance of GBV research. Community oversight and feedback during project implementation are important tools for monitoring risks and identifying program adaptations to reduce the risks and enhance the impacts of women’s empowerment. 5.2 ANALYZING THE PREVALENCE OF GBV AND KEY RISK FACTORS Information on the prevalence of IPV and nonpartner violence and on attitudes toward IPV usually is available through national Demographic and Health Surveys and Multiple Indicator Cluster Surveys. These data are disaggregated by income level and regions and thus can provide information relevant to specific beneficiary populations. An overview of these and additional data sources is provided in Table 1. 68 In FCV and other emergency situations, for which the environmental and social management framework/plan can be prepared after appraisal, mitigation measures still need to be established before project activities start. 22 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE TABLE 1. Sources for obtaining GBV-related information TYPE OF INFORMATION SOURCE Prevalence of GBV • Demographic and Health Surveys • Multiple Indicator Cluster Surveys • Quantitative and qualitative studies GBV context • Qualitative data from NGOs (attitudes, norms, legal context) • Local women’s organizations and key informants • Demographic and Health Surveys • International Men and Gender Equality Survey • Quantitative and qualitative studies • Policy and legal context data • Participatory data collection activities Services available for GBV survivors • Qualitative data from NGOs • Local women’s organizations and key informants • Resource mapping activities Given the numerous ethical and safety concerns related to the collection of GBV prevalence data, these data should not normally be collected by project teams during risk assessment or as part of regular risk monitoring activities. A lot of information on gender dynamics in households and communities more generally will also be available from previous gender assessments by World Bank teams or other development stakeholders, such as UN Women, local nongovernmental organizations, or national counterparts. However, new analysis will be required to identify the interplay between SSN design features and GBV dynamics within beneficiary households and communities. This information should be collected as part of the environmental and social assessment. However, in practice, the assessment is not always completed in time to influence project design because of capacity constraints, tight timelines, and other problems. In such cases, teams should consider commissioning additional analytical work on gender and GBV issues that are relevant to project design. Overall, violence is more prevalent among specific at-risks groups that experience multiple forms of vulnerability. Table 2 identifies at-risks groups that might require special initiatives to reduce GBV risks. In Cameroon, the likelihood that cash transfers will contribute to either increased or decreased risk is based largely on the intrahousehold dynamics. Violence was more likely if the male heads of household previously exhibited controlling behaviors or expressed negative attitudes toward women’s empowerment or where there was substance abuse.69 This underlines the importance of engaging men in efforts to improve attitudes toward gender equality. However, factors linked to a greater likelihood of violence may contribute to increased risk, but not be direct causes. Not everyone who has been identified as at risk becomes a victim of violence. 69 Kuttner, Ngo Likeng, Schuettler, and Yossa. (2020). The environmental and social framework 23 TABLE 2. Key sources of vulnerability among at-risk groups AT-RISK GROUP FACTORS THAT CONTRIBUTE TO INCREASED RISK OF VIOLENCE Adolescent girls • Increased domestic responsibilities that keep girls isolated in the home • Lack of access to clear information about health care and services, including GBV survivor support services • Individuals are discouraged or prevented from attending school • Early pregnancies and motherhood • Child marriage • Dependence on exploitative or unhealthy relationships for basic needs Elderly women • Weakened physical status, physical or sensory disabilities, and chronic diseases • Isolation and higher risk of poverty • Risks of manipulation or exploitation by adult children • Limited mobility • Lack of access to clear information about rights and services Woman and child heads • Increased domestic responsibilities that keep them isolated in the home of household • Erosion of normal community structures of support and protection • Dependence on exploitative or unhealthy relationships for basic needs Indigenous women, girls, • Social stigma and isolation men, boys, and ethnic and • Poverty, malnutrition, and reproductive health problems religious minorities • Lack of protection under the law and high levels of impunity for crimes against them • Lack of opportunities and marginalization based on their national, religious, linguistic, or cultural group • Barriers to participating in their communities and earning livelihoods Women, girls, men, and boys • Limited mobility, hearing, or vision resulting in greater reliance on assistance with disabilities and care from others • Isolation and a lack of social support or peer networks • Exclusion from sources of information and guidance because of physical, technological, and communication barriers • Physical, communication, and attitudinal barriers in reporting violence Refugees and internally • Elevated levels of insecurity and GBV in affected communities and camps for displaced persons refugees and internally displaced persons • Increased stress within displaced households leading to increased risk of violence against women, children, and other dependents • Vulnerability to sexual and other forms of exploitation by service providers and other program actors, given the heightened dependence on humanitarian assistance Source: Adapted from IASC 2015. 24 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Table 3 includes key information that is relevant for understanding these factors. Much of the information on context and the legal framework will be collected as part of the SEA/SH risk assessment. In most countries, social and gender experts will have readily available gender analysis covering most of these elements used throughout the country portfolio. TABLE 3. Example of a GBV analysis framework THEME KEY INFORMATION Country context • Prevalence of IPV; prevalence of any form of sexual violence; attitudes toward domestic violence disaggregated by wealth quintile and region (Demographic and Health Surveys and Multiple Indicator Cluster Surveys) • Gender and GBV dimensions of FCV-affected or humanitarian situations in areas in which program implementation will take place Legal context • Legal framework on physical and sexual assault and harassment, marital rape, minimum age at marriage and divorce, inheritance rights, women’s independent or joint land ownership, women’s independent access to financial services, labor laws, and so on • Traditional and religious laws and cultural norms on gender-based violence, marriage, divorce, restrictions on women’s mobility, inheritance, and so on Survivor support services* • National systems and referral pathways for the provision of support services to GBV survivors • Community-based support networks and warm referral pathways Labor markets (for public works • Female formal and informal labor force participation rates and constraints and livelihoods programs) • Social and legal norms regulating women’s formal or informal economic activity Household composition, decision- • Control over and division of household resources and any variation based making, and resource allocation on household structures (woman-headed and man-headed households, monogamous and polygamous households, multigenerational households, and so on) • The intrahousehold decision-making process (individual and joint decision- making, spheres of control, and so on) across different types of household structures • Gender division of labor and care responsibilities * This information is required to establish a grievance mechanism that can respond to GBV-related complaints. Ideally, the mapping exercise will be completed by the country team rather than through individual projects. Another important consideration is the heterogeneity across beneficiary households relative to factors that can either increase or reduce risks within the household. Certain characteristics or dynamics within households may raise the risk that a SSN program triggers or exacerbates GBV.70 Factors may include the dependency ratio within the household, the relative educational attainment of intimate partners, the employment status of adult household members, the type of household structure, and so on. Generally, data will be available on such socioeconomic characteristics of intended beneficiary households and their members, primarily through household surveys, and teams may then locate evidence linking 70 Data on 21 countries suggest that the differential educational attainment of partners may play a key role in predicting IPV risk. A family history of violence, attitudes toward wife beating, early marriage, polygamous marriage, and a husband’s alcohol abuse are the most meaningful risk factors. Education is a key protective factor against GBV (World Bank 2014). The environmental and social framework 25 characteristics to the prevalence of violence.71 On this basis, teams may be able to determine whether the program should include modifications or accompanying measures targeted at subgroups of beneficiaries or monitor the impact of the program on certain groups of households more closely. The risk of GBV is often intensified during shocks and in FCV contexts because of heightened insecurity, the instability of social structures, a lack of access to services, and increased stress in the affected households. Given heightened levels of insecurity, women and girls often face greater risks of opportunistic harassment or assault by security forces and persons other than intimate partners or project actors if they participate in program activities or must travel to collect transfers. The presence of security forces and militia often heightens this risk. Higher levels of stress in households experiencing shocks and insecurity can also lead to conflicts and a rise in IPV. Beneficiaries could be at greater risk of SEA if they are dependent on program resources. Compounding this situation, medical, social, and other support services are often weak or nonexistent, leaving GBV survivors without support.72 It is therefore important to consider the particular risks related to FCV during the risk assessment process and adopt appropriate mitigation measures for these contexts. The specific issues to be investigated will depend on the type of SSN program being delivered. For example, the question of whether, on average, households with male or female primary income earners are poorer may be a key consideration in the design of a targeting strategy for poverty-focused SSNs. For food security interventions, identifying which household member is usually responsible for food purchasing and how food is distributed within the household is important. Identifying the main gender gaps in access to income-generating activities—occupational segregation, wage gaps, and access to productive assets and financing—are important in designing livelihoods interventions. Understanding the gender dimension of household investments in human capital is important in programs focused on human development. These contextual factors can either drive or mitigate GBV risks and are thus relevant to SSN design choices. (See Annex 2). 71 See Kishor and Johnson (2004) for an example. 72 See Cross, Manell, and Megevand (2018) for more on cash transfers in humanitarian settings. 26 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 6 ASSESS 6.1 OUTREACH73 Outreach involves interactions to inform people about social protection programs and delivery processes to encourage engagement. It is the first phase of the delivery chain and is critical to the effectiveness of any social protection program. While continuous communication and outreach are important throughout the delivery chain, this subsection focuses on initial outreach, which aims to inform the intended and wider population about programs to help them become aware, informed, able, and encouraged to engage. Effective outreach lays a foundation for transparency and an understanding of the program by all stakeholders, which can contribute to the prevention of backlash against women beneficiaries and can strengthen household and community buy-in for program participation. KEY QUESTIONS „ Can program objectives explicitly recognize women’s empowerment and gender equality without risk of male backlash? If not, can objectives be framed in a way that nudge beneficiary households toward greater gender equality (for example, by linking program objectives to household well-being that requires both men and women to play a rolewithout any explicit reference to equality or empowerment)?74 „ How can program objectives, benefits, and processes be communicated in a way that promotes women’s participation, while avoiding risks of backlash or risks of reinforcing traditional gender norms? „ What communication channels are most effective in reaching women, men, community leaders, or other norm influencers (such as mothers-in-law and elders) with key information about the program? „ What constraints and accessibility issues do women face in participating in outreach activities and how can these be reduced? What accommodations or adaptations should be made during the communication process to respond most effectively to the needs of women? „ Are there opportunities for involving women’s groups or other local structures that already engage women in outreach activities? Outreach is a key step in building support and securing the participation of the population. SSNs typically develop an outreach strategy that combines multiple modalities, including (a) direct outreach to potential beneficiaries by outreach officers; (b) community-based outreach, such as through community leaders or women’s groups; (c) outreach through referrals from community organizations, service providers, or other 73 The purple boxes at the beginning of each subsection are based on Lindert et al. (2020). 74 For more on nudge theory in behavioral economics, see Bohnet (2016). Assess 27 programs; and (d) indirect outreach through print, social, or mass media.75 In the design and implementation of outreach activities, several factors should be taken into account to reduce the risk of GBV and promote women’s participation (Good Practice Box 4 ). GOOD PRACTICE BOX 4. DESIGNING A PROGRAM COMMUNICATION STRATEGY A program communication strategy should involve steps to ensure that girls and women are informed and consulted throughout the program cycle. There are at least five key components of any communication activity, whether consultation, outreach, behavior change communication, or feedback activity. At each stage, it is possible to strengthen women’s inclusion and build support for women’s participation. „ Identify and assess the target audience, including determining who, in addition to women, are important gatekeepers and influencers in the community (for example, women leaders, community elders) „ Develop the content or key messages to be communicated, including developing messages that address the cultural context „ Identify or establish the channels through which messages will be communicated, including identifying multiple entry points to potential beneficiaries and local information gatekeepers and norm influencers „ Deliver key messages, including the repetition and reinforcement of messages to multiple audiences „ Solicit continuous feedback and making adjustments where necessary, including the establishment of consultation mechanisms that create safe spaces allowing women to communicate any concerns or suggestions to mitigate emerging risks. All messages should be formulated with reference to the social and cultural context. In some cases, it will be possible to promote women’s empowerment and GBV prevention as an explicit program objective. In other settings, emphasizing the positive impacts on a broader population of households may facilitate women’s participation and reduce the risk of male backlash. Men may be less likely to react negatively if women’s participation is not seen to challenge their role as household head and primary breadwinner. At the same time, it is important to try to avoid reenforcing traditional gender norms on the division of responsibilities and labor, such as women’s roles as primary caregivers. Involving local stakeholders and influencers is an effective means to build buy-in for programs, particularly if women’s participation challenges traditional gender norms, for example, through their engagement in nontraditional public works or traveling outside their communities (Evidence Box 5). 75 See Lindert et al. (2020), Table 3.2. 28 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE EVIDENCE BOX 5. COMMUNICATING PROGRAM OBJECTIVES AND LABELING How program objectives are communicated can influence the impact of a transfer on gender dynamics and IPV as much as the fact that women are the designated transfer recipients. For example, evidence from Colombia and Ecuador shows that more generous transfers to women were accepted by men and did not lead to increased conflict or violence in part because the interventions were clearly communicated as intended for food and nutrition, a domain already considered women’s responsibility and thus nonthreatening to men’s status.a Some programs have experimented with the idea of labeling, that is, explicitly communicating the purpose of the transfer and, by implication, how it should be spent, but without enforcing any conditions. Evidence from Kenya suggests that labeling can result in similar perceptions of the program rules and expectations between households that are subject to conditions and households that are not.b An evaluation in Morocco found that a cash transfer without conditions, but explicitly labeled as an education support program led to large gains in school participation.c The idea of labeling could be tested to promote more equal gender roles, for instance, by recognizing the value of and engaging men in care work or joint household budgeting. a. Hidrobo, Peterman, and Heise (2016). b. Heinrich and Knowles (2020). c. Benhassine et al. (2015). Women often face constraints on accessing information or participating in outreach activities. The constraints can be the result of various factors, including limitations on women’s mobility (because of social norms, security risks, and so on), time constraints because of care work, difficulties understanding communication language and formats (for example, if women have poor literacy or if information is not provided in local languages), and so on. Some women, such as women with disabilities or members of minority groups, may face additional difficulties and require adaptations. Women’s groups and organizations can be an effective communication channel. Outreach is an important opportunity not only to ensure that women are informed about a program, but also to build support for women’s active participation. By including women in outreach activities, a program signals that women’s participation is important, which can help reduce potential backlash. At the community level, programs intending to serve girls and women should involve traditional and religious leaders early in the design and delivery process to obtain their support and, ideally, enlist them as allies to facilitate women’s participation. In some settings, women and girls depend on gatekeepers (for instance, fathers, husbands, or mothers-in-law) in accessing information and services or for permission to travel and work outside the home. Reaching out to communities to ensure buy-in of women’s participation is often an iterative process that requires, particularly in more conservative settings, multiple consultations with those who control local norms and practices (such as community elders and other traditional leaders). Furthermore, in conservative contexts where men act as the gatekeepers, strategies may be needed to ensure that they pass the relevant information on to women (Good Practice Box 6). Assess 29 GOOD PRACTICE BOX 6. DESIGNING AN OUTREACH STRATEGY IN A CONSERVATIVE FCV CONTEXT Designing an outreach strategy can be particularly challenging in conservative societies characterized by strong patriarchal norms in which women’s participation in the public sphere is constrained. Leveraging traditionally acceptable channels of communication can help generate support from traditional and religious leaders for women’s participation in an SSN program and reduce the risk of backlash. The Temporarily Displaced Persons Emergency Relief Program in Pakistan’s former Federally Administered Tribal Areas adopted a tactical gender- and conflict- sensitive communication approach capitalizing on local knowledge with reference to scriptural sources and local cultural norms to sensitize community elders and imams. These leaders then acted as interlocutors and mobilizers of the community to support and encourage high levels of participation among women. The larger program delivery strategy was also tailored to respect local norms and practices and increase women’s access to benefits and services. Examples include the following: „ Recruitment of local community mobilizers, particularly women, was key to ensuring women’s participation in the program. To address the perceived need for male relatives to act as interlocutors in communication with women, around 50 percent of the mobilizers recruited were women. This allowed the program access to the domestic sphere where local women could be reached directly to ensure their voices were heard and their distinct needs and priorities were addressed. „ Men members of beneficiary households were sensitized separately on how the participation of women ensures greater improvement in the well-being of the entire household and community. „ Because women were required to attend child health and nutrition awareness sessions to receive child wellness grants, the sessions were organized for women-only groups to provide comfortable spaces for sensitization and interaction. „ The program facilitated women’s civil registration at citizen facilitation centers at which child health and nutrition awareness sessions were conducted to allow the women to receive the cash transfers. Lacking national identity cards, women had been unable to complete the mandatory biometric identity verification process and thus needed to be accompanied by a household member in possession of a national identity card, usually their husband. „ Adaptations were introduced at the citizen facilitation centers to reflect local cultural norms and ease women’s access, including the adoption of service delivery timetables with women-only days, the designation of special women’s counters, the erection of privacy barriers in waiting areas, and so on. „ Significant efforts were undertaken to recruit women front-line service providers, including the provision of safe transport and accommodation for women program workers. However, this remains a challenge as the program expands into new districts. DESIGN AND IMPLEMENTATION TIPS „ Undertake diagnostics to identify sensitivities around the program’s development objectives and women’s involvement in program activities. „ Consider framing objectives related to women’s economic inclusion, the designation of transfer recipients, or participation in accompanying measures as beneficial to households and the community (such as reducing household poverty, increasing household income, improving children’s health and schooling, and so on). 30 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE „ In contexts in which women’s access to information is limited, consider working with information gatekeepers to facilitate the transmission of key messages to women and work to increase the acceptance of women’s direct access to information. „ In planning communication and feedback activities, implement adaptations to facilitate women’s effective participation in or access to these activities, taking into account literacy levels, minority languages, time and mobility constraints, social constraints, security concerns, and so on. „ In selecting communication channels, assess the ability of the channels to reach various groups of women effectively, including by taking intersecting vulnerabilities, such as disability, into account. „ Monitor outreach activities to ensure these groups are effectively reached and adjust or expand the outreach strategy as needed. „ Provide information in multiple formats, including written, oral, and easy to read formats to improve accessibility. To the extent possible, communicate in local languages. „ Ensure that beneficiaries and other household members understand program requirements, such as participation in accompanying measures, training programs, or public works, to reduce the risk of backlash within households against beneficiaries. INTAKE, REGISTRATION, AND THE ASSESSMENT OF NEEDS 6.2  AND CONDITIONS Intake, registration, and the assessment of needs and conditions are the second and third phases of the delivery chain; their aim is the efficient registration of the target populations and the accurate recording of personal information and the profiles of participants. Intake is the process of initiating contact, engaging clients, and gathering information, while registration consists of recording and verifying the information; the two steps are usually simultaneous. The assessment of needs and conditions is the process of profiling registered individuals or households according to various assessment tools. In low-income countries, SSNs, particularly cash or in-kind transfers, predominantly use administrator-driven approaches, such as categorical targeting, proxy-means testing (PMT) or community-based targeting (CBT), to gather information and profile households. On-demand approaches, such as self-registration, are more commonly used for employment- support interventions (public works, livelihoods interventions) or for programs in higher-income settings. These processes can be conducted separately for different programs (such as cash transfers and public works) or can be combined, and the assignment of benefits can be determined based on the profiles identified. KEY QUESTIONS „ What are the constraints and opportunities to increase women’s effective inclusion in the processes of intake, registration, and the assessment of needs and conditions? If they exist, how can any constraints associated with requirements to present legal proof of identity be reduced? „ What are the risks of SEA—for example, a request for sexual or other favors in return for inclusion in the program— by program actors associated with different intake, registration, and assessment activities, and what measures can be taken to reduce and mitigate these risks? Assess 31 Intake and registration Intake may be initiated either by the applicants themselves or by program administrators, which has implications for women’s inclusion in the program and for the exposure to risks of SEA and other abuses of power. Administrator-driven intake is typically carried out at the household level, usually through door-to-door registration campaigns or on the basis of information in existing social registries. It is the most common intake and registration method in SSN programs in lower-income countries. En masse registration drives, also known as census sweeps, are often administered in areas with a high concentration of potential beneficiaries (that is, high poverty rates). Door-to-door registration is intended to reduce exclusion in these contexts, though registration can also be centralized in community centers or other local venues. However, as in any face-to-face interaction whereby individuals have the power to affect another person’s access to benefits, face-to-face registration can raise the risk of sexual or other forms of exploitation or abuse. Separating the process of registration and the individuals involved from the process of the assessment of needs and decisions regarding eligibility (and the individuals involved) can help minimize the risk of such abuses of power. If feasible, blinding the registration agents to the eligibility criteria and restricting their role to the collection of information may substantially reduce the risk of manipulation or other abuses of power. To limit further the opportunity for the manipulation of the process, the decision to register a household or individual should be based on clearly established objective criteria and, if possible and relevant, automated. Given resource and capacity constraints, few census sweeps cover the entire population, and any listing or preregistration may create opportunities for SEA if transparency and oversight are not adequate. Instead, programs often use quotas to limit the number of households that can be registered and may rely on communities themselves to prioritize which households are registered either through consultation with local leaders or through a community-based selection process. While useful in leveraging local knowledge and applying more locally relevant poverty metrics, decentralized registration can create opportunities for the abuse of power by those responsible, who may seek sexual or other favors in exchange for inclusion. If feasible and appropriate, public and participatory meetings should be held to validate the list of potential beneficiaries and limit abuses of power and opportunities for exploitation. The list of registered individuals or households can also be made public following door-to-door data collection to minimize opportunities for registration agents to use the threat of omission as a way to extract sexual favors or bribes, though risks of stigmatization in some contexts might disqualify this approach. There may also be practical challenges to registering women because of lower rates of national ID coverage or mobility constraints. Lack of possession of national IDs is a common barrier, particularly among women and especially in Sub-Saharan Africa and South Asia. In places where civil registration rates are low, SSN programs aiming to register women need to address the issue either by allowing alternative forms of identification, such as recipient or client cards, or by facilitating civil registration in parallel with the registration process.76 Mobility constraints related to restrictive social norms, security risks, or unequal access to transportation may prevent potential women beneficiaries from registering. Robust recruitment, training, and supervision mechanisms for those in charge of these processes— coupled with clear and enforceable codes of conduct—can reduce and mitigate the risk of abuses of power during administrator-driven registration, especially if the processes involve home visits.  egistration campaigns were organized with the authorities responsible for civil registration to provide ID cards to beneficiaries in the Girls’ 76 R Education and Women’s Empowerment and Livelihoods Project (GEWEL) in Zambia and the Temporarily Displaced Persons Emergency Relief Program in Pakistan. Mauritania’s Social Safety Net Project II is supporting with the civil registration among beneficiary household members as part of the government’s expansion of universal access to health care. 32 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Whether data collection is performed by program staff or other civil servants or is outsourced to survey firms or local NGOs, it is crucial that the program establish clear codes of conduct and incorporate SEA prevention into the training and supervision of registration agents (see Section 9.1 Codes of conduct). Supervisors should receive additional training on accountability and reporting mechanisms and be made responsible for reporting any malfeasance. Crucially, a grievance mechanism with established protocols and the capacity to receive SEA/SH complaints must be active during intake and registration (See Section 9.2 Grievance mechanisms). On-demand registration is commonly used in more developed social protection systems, especially for employment-support interventions. Eligible individuals are invited to self-register through public information campaigns. Enrollment in programs is then offered to those verified as eligible on a first- come, first-served basis, through lotteries, or by using additional program criteria to prioritize among the eligible (for instance, age or gender). Communication, particularly outreach, is critical to ensuring that the process is fair, transparent, and inclusive of women and other marginalized groups, especially those with lower levels of digital literacy or less access to registration technology. Quotas on women’s participation are frequently used, in conjunction with outreach, to ensure that women are adequately served. Potential SEA risks as a result of abuses of power by the program staff conducting the outreach or processing the intake and registration can be minimized through clear and transparent processes and robust supervision. Regardless of the intake and registration approach, special attention is required to minimize exclusion or other unintended adverse effects on women’s participation in expediting processes in response to a shock. Governments are increasingly investing in adaptive social protection systems and in national social registries to facilitate the rapid extension of coverage in response to shocks, such as natural disasters, pandemics, or economic shocks or other shocks arising from human activity. The COVID-19 pandemic has demonstrated that SSN programs can provide an important framework for rapid shock response. However, in fast-tracking registration to scale up quickly, it is important to identify underlying constraints (such as gender gaps in mobile phone ownership or legal proof necessary for IDs) that may affect women disproportionately and put in place appropriate countermeasures. Assessment of needs and conditions The assessment of needs and conditions involves systematic processes for profiling registered individuals, families, or households according to various assessment tools. The instruments and techniques for assessing needs and conditions vary depending on the characteristics of the target group. One method is simply to classify applicants according to demographic characteristics, such as gender or age (in the case of demographic categorical programs). Another approach relies on caseworker assessments (common in employment and social service assessments). A third method involves the use of automated formula to aggregate key indicators (typically used for aggregating socioeconomic welfare measures). This subsection focuses on two assessment approaches often used sometimes in combination (and often combined with geographic targeting) to profile households for poverty-targeted SSN programs in low- income countries. „ Proxy-means testing (PMT), which estimates a household’s socioeconomic welfare using a composite measure that estimates welfare as an index based on a weighted score of observable household characteristics. The word proxy reflects the fact that observable characteristics are considered proxies for actual income or consumption. Assess 33 „ Community-based targeting (CBT) whereby communities (through representatives or a participatory process) assess the needs and conditions of households by classifying them according to socioeconomic status or verifying compliance using categorical criteria. CBT can also be used in combination with PMT to prioritize which households will be registered and then subject to a PMT or to help validate PMT estimates. From a gender perspective, there are advantages and disadvantages to both PMT and CBT targeting approaches. PMT assessments of socioeconomic status are typically conducted at the household level rather than at the individual level, thereby obscuring intrahousehold differences that tend to disadvantage girls and women. In polygamous households, depending on the assessment unit, household-level assessments may overlook inequalities between the different cells or units associated with each co-wife and her dependents. At the same time, while a PMT approach is less susceptible to local biases, it can often miss key unobservable aspects of household poverty and women’s vulnerabilities, such as women’s limited control over household resources. CBT, on the other hand, may be more sensitive to local poverty dynamics, but more prone to elite capture; it might lead to inequitable treatment across areas or other forms of bias that can disadvantage or exclude women. Through intentional program design, both approaches can be leveraged to amplify women’s voices and improve assessment accuracy and inclusion. Administering data collection tools to male heads of household leads to answers that are different from those women would give, which may be more accurate. While women’s knowledge of household assets and consumption varies greatly across contexts, intentionally engaging women in data collection may make the process more inclusive and equitable, while also improving the quality and accuracy of the information collected. Reliance on the concept of a household head as the organizing principle for surveys may also need to be reevaluated and, where appropriate, replaced with individual-level data collection and analysis to capture intrahousehold inequality. CBT can similarly be leveraged to provide a platform for women’s voices and enable women to influence the program intake, registration, and assessment processes, ensuring that gender-specific vulnerabilities and local poverty dynamics are taken into account. Implementers should not assume women’s participation would be achieved simply by conducting public community meetings. Rather, they should establish special measures, if relevant, such as conducting separate meetings with women, to address social norms that may prevent women from speaking up in public if their husbands or other male community members are present. As with intake and registration, the risks of SEA and other abuses of power may emerge during the process of assessing beneficiary needs and conditions. PMT approaches typically included face-to- face interactions between enumerators and potential beneficiaries whereby enumerators may have the power to impact whether households or individuals can access program benefits. It is thus important to disseminate and train supervisors and enumerators on codes of conduct that address SEA (see Section 9.1 Codes of conduct). CBT approaches can also create opportunities for abuses of power, including SEA, by allowing individual community representatives to determine whether households qualify for inclusion in a program. A participatory CBT process, given the collective nature, is less susceptible to individual abuses of power. It is important that communities be made aware of the process for registering complaints through the program grievance mechanism of any abuses of power during the assessment of needs and conditions (see Section 9.2 Grievance mechanisms). Similar SEA risks and mitigation strategies apply to the recertification or reassessment and exiting phases of a social protection program. Most SSN programs provide benefits and services to beneficiaries for a predetermined time period, and some also institute recertification or reassessment to decide who 34 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE should stay in the program beyond this time period. The reassessment process should take into account the guidance offered for the initial assessment of needs. Similar guidance applies to programs that have explicit processes to exclude some households. DESIGN AND IMPLEMENTATION TIPS „ Engage communities or local representatives, including women leaders, in the registration process to learn about perceptions of poverty and collect feedback on selection criteria and the approach used. If communities are less well organized or less cohesive (for example, in urban areas), the involvement of community-based organizations in the validation process may be an effective strategy. „ Invest time and resources in the recruitment, training, and supervision of the individuals responsible for intake and registration to ensure that only those qualified and vetted are part of the process, whether managed directly or through a third party. „ Consider mandating gender quotas for intake and registration staff, including for those on selection and appeals committees. „ To the extent possible, deploy registration agents who reflect the profile of the assessment sample (for example, in sex, ethnic, or linguistic group) and understand the community to facilitate access and increase the accuracy of the data collected, while ensuring the robust monitoring of any local bias toward potential registrants. Facilitate the recruitment of women agents, including by ensuring SH mitigation measures are in place and by taking steps to minimize the risk of the opportunistic harassment or assault of women agents engaged in the registration process. „ Provide training on codes of conduct that prohibit sexual and other forms of exploitation among all persons involved in the intake, registration, and assessment of needs and conditions. „ To limit the potential for manipulation, avoid disclosing to enumerators any information on the indicators used during the process of the assessment of needs. „ Adopt special measures to ensure that the at-risk groups (identified during environmental and social assessment) are identified and included during intake and registration and during the assessment phases. „ Separate the process and the individuals involved in information gathering from decision-making on eligibility (at the next stage—enrollment). „ Where relevant, avoid imposing conditions for registration (for instance, presentation of an ID) that could affect women disproportionately (or offer alternative methods of identification). „ To ensure participation, organize activities while taking into account women’s potentially constrained time and mobility. Assess 35 7 ENROLL The enrollment stage involves determining who qualifies for programs and what benefits and services they will receive. The objective is to determine eligibility effectively according to specific criteria, confirm that the benefit and service packages are accurately determined, and ensure that eligible registrants are enrolled and onboarded efficiently without errors of exclusion or inclusion. Once people’s needs and conditions have been assessed, their profiles are compared to eligibility criteria for specific programs. Three key elements help determine the eligibility for social protection programs: eligibility criteria, profiles of registrants based on assessments of needs and conditions, and the definition of the beneficiary unit (individual, family, or household). The intended population enters this stage as registrants and becomes beneficiaries if they are deemed eligible, enrolled, and onboarded. KEY QUESTIONS „ What are the opportunities associated with designating women as the recipients of SSN benefits? Would transferring resources to women contribute to promoting women’s empowerment and changing norms? What are the estimated risks of backlash in the immediate and longer term and how can these be mitigated? „ Is there an instrumental reason for designating women as the recipients of SSN benefits, such as improving outcomes among children? Can this be used to strengthen women’s position as valued members of the household? „ What barriers and constraints do women face in enrollment? What mechanisms can be used to facilitate women’s enrollment? „ Are there any proof of identity or other requirements that limit women’s ability to be enrolled? How can these be modified to allow for more inclusive provision of assistance or are there opportunities for the program to contribute to closing gender gaps in access to IDs? Several factors beyond the consideration of women’s empowerment and GBV prevention are involved in decisions on eligibility and enrollment, depending on program objectives and context. For example, SSNs may prioritize the elderly or persons with disabilities if the objective is to provide income support to at-risk groups; they may prioritize households with labor capacity if the aim is to provide temporary wage-labor opportunities; or they may prioritize pregnant women or mothers of young children if early childhood development is a key goal. Some programs may prioritize the enrollment of woman-headed households if these households are assessed as poorer and more vulnerable. In addition to deciding who should be a beneficiary, the program must designate the person who will be the recipient of services or benefits (and, at times, an alternate). If programs aim to support all members of households (for example by reducing household poverty or food security), the benefit recipient will be de facto acting on behalf of the household. In instances in which the sex of the household head is not a factor in determining a household’s eligibility, programs may nonetheless specify the sex of the benefit recipient. Women are 36 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE often designated as recipients because of the role they play in driving consumption and human capital outcomes. Empirical evidence on the differential use of transfers based on the sex of the recipient is mixed; more research is needed. (Evidence Box 7). While all SSNs have the potential to reduce violence in beneficiary households by lessening poverty- related stress and conflict over scarce resources, designating women as transfer recipients may have additional GBV impacts. Transferring resources directly to women has the potential to lower IPV rates by empowering women. If women can retain control of these resources, their dependence on men partners will be diminished, improving their ability to makes bargains over violence. However, contingent on various factors at the individual, household, and context level, empowering women through resource transfers may also lead to a backlash that can become manifest as violence if the men partners perceive that their status or authority is being undermined. In some cases, programs engage women in income-generating activities through public works or livelihood grants, and there may also be community backlash against any perceived transgression of gender norms on women’s work. These risks can be reduced and opportunities for GBV prevention can be maximized through design choices. For instance, engaging all members of households in communication on program objectives and the intended use of transfers may help prevent conflicts over the use of the resources. In the Social Safety Net Project in Cameroon, household heads who are predominantly men, are required to sign a moral contract whereby they agree to respect all program requirements, including the participation of their partners in program activities and ensuring that transfers are spent in accordance with program objectives. This experience could be adapted to other settings. Spouses could jointly sign a moral contract to establish shared responsibility over the use of program benefits and participation in program activities.77 Similarly, in Ethiopia’s Productive Safety Net Program, client cards list the names of heads of household and spouses. Additionally, across contexts, engaging local leaders as well as men and boys to build support for women’s economic activities is important in lessening the potential for backlash for any perceived transgression of gender norms. The decision tree in Figure 5 presents the opportunities, risks, and mitigation measures associated with various enrollment decisions related to the sex of the recipients of SSN benefits and services. 77 Kuttner, Ngo Likeng, Schuettler, and Yossa. (2020). Enroll 37 EVIDENCE BOX 7. SEX OF TRANSFER RECIPIENTS Cash transfer programs typically designate women as transfer recipients because of the associated benefits for households. A body of research shows that individuals in households may have different preferences and abilities to bargain to realize their preferences. For instance, research in Brazil, Côte d’Ivoire, and, more recently, Ghana finds that changes in the control over income among individual household members are associated with changes in expenditure patterns and, specifically, that increasing income among women leads to greater spending on food, schooling, and health care. Similarly, women’s bargaining power has been associated with better educational and health outcomes among children. Cash transfer programs, particularly in Latin America, often designate women as direct transfer recipients to maximize the impacts on the outcomes among children and the overall well- being of the families. More recently, programs have begun shifting away from targeting women instrumentally in pursuit of better human development outcomes among children to designating women as recipients to improve the financial inclusion and economic empowerment of the women. Despite the link between women’s earned income and outcomes among children, there are no systematic differences between men and women recipients. A 2012 systematic review documented sex-differentiated impacts on outcomes in 13 of the 15 studies examined.a However, while the transfers to women appeared to improve child nutrition and health, there was little evidence that targeting women led to other systematic changes in economic choices or to better household benefits relative to the results of transfers to men. The findings of a more recent review similarly find no conclusive evidence on the role of cash transfers to women versus men.b Cash transfer evaluations in Burkina Faso, Kenya, and Morocco likewise find no differences.c A recent review of 35 studies found that providing women with access to financial resources or services through microcredit, savings accounts, or cash transfers did not consistently enhance women’s economic empowerment.d Empirical evidence on the effect of a transfer recipient’s sex on IPV outcomes is limited. A study in Kenya that directly compared men and women transfer recipients found that the decline in IPV was significant regardless of the sex of the recipient, but that the magnitude of the effect was larger if the transfers were directed toward women. The authors concluded that their results were broadly consistent with the view that transfers to women rather than men boosts the bargaining power of the women.e Evidence on a program in Mali in which men were the recipients of cash transfers showed a reduction in physical IPV, although no difference in terms of sexual violence or controlling behavior.f Because these transfers to men did not lead to increases in women’s agency to challenge men’s authority or bargaining power in sexual or financial matters, the authors suggest that making women the designated recipients, combined with skills and knowledge building, would be more effective in promoting equality and preventing violence. More evidence is needed, particularly on more conservative settings in which the risk of a backlash after the selection of women as transfer recipients may be greater. a. Yoong, Rabinovich, and Diepeveen (2012). b. Bastagli et al. (2016). c. Akresh, de Walque, and Kazianga (2013); Benhassine et al. (2015); Haushofer and Shapiro (2018). d. J-PAL (2021). e. Haushofer and Shapiro (2018). f. Lees et al. (2021). As in the case of registration, there may be practical challenges to enrolling women as SSN recipients. Their lack of national IDs, lower literacy levels, and more limited mobility, either culturally mandated or because of unequal access to means of transportation and resources, may hinder women’s enrollment as recipients of SSN programs and increase their vulnerability to GBV. 38 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE FIGURE 5. Choosing beneficiaries and recipients of benefits and services Is gender an aspect of household or individual eligibility? NO YES All households are eligible Woman-headed households are prioritized (if beneficiary = household) (if beneficiary = household) or or Both men and women are eligible Women are prioritized (if beneficiary = individual) (if beneficiary = individual) Opportunities Risks • Reduce poverty or food • Community backlash or insecurity among vulnerable stigmatization of vulnerable women women or woman-headed or woman-headed households households • Use of violence or threats of violence • Provide income support, by family to extract resources income generation, or wage • Traveling to or from activities or labor to vulnerable women or during participation in program woman-headed households activities Are women YES designated as the NO default recipient? Mitigation • E ective communication strategy • Safety precautions • Involve men in accompanying measures Women designated Household selects recipient as recipients (often male household heads) Opportunities Risks Opportunities Risks • Women’s economic • Man’s backlash if woman • Reduce risk of violent • Missed opportunity to reduce empowerment and IPV is perceived to challenge male backlash women’s financial dependence reduction his role as breadwinner on head of household • Resources more likely to • Confiscation of resources • Missed opportunity to be spent toward human • Risks of opportunistic GBV increase women’s bargaining development outcomes traveling to or from power • Increase women’s activities or during • Reinforces women’s bargaining power in the participation in program dependency on men for household activities financial support. • Shift gender norms that dictate women and men’s work and livelihoods, for example, through labeling • Potential opportunity to Mitigation Mitigation increase women’s financial inclusion or literacy • Soft labeling and • Soft labeling and communication strategy communication strategy • Prior household agreement • Prior household agreement on use of resources on use of resources • Safety precautions • Engage women in human • Involve men in development or other accompanying measures accompanying measures Enroll 39 DESIGN AND IMPLEMENTATION TIPS All programs „ Gather data on household decision-making for the allocation of resources and on how transferring program benefits to women could enhance women’s bargaining power. „ In designating recipient, evaluate whether providing benefits and services directly to women could contribute to women’s empowerment, given the social and cultural constraints. „ Clearly communicate the intended use of program services or benefits (for example, by labeling; see Evidence Box 4), and, if possible, engage beneficiary households in joint planning for the use of program services or benefits, for instance, through household budget planning or joint signature of moral contracts by intimate partners specifying they will use transfers for the intended purposes. „ In enrolling polygamous or nonnuclear households, identify the recipients most likely to allocate program resources for the intended purposes as follows: ƒ Allow households to select a single recipient if the intended use of resources is within their purview and traditionally distributed equitably among household units (for example, food transfers in a context in which food preparation is a collective responsibility and households eat from a single pot). ƒ Explore the possibility of enrolling multiple wives separately if program resources are unlikely to be distributed equitably among household units (if co-wives are considered part of the same household). Programs designating women as primary recipients of benefits or services (in addition to the above) „ Identify any risks of backlash in beneficiary households, risks of opportunistic harassment or assault if engaged in program activities outside the household, and risks of SEA while interacting with program actors (see Section 6 Assess and Section 8 Provide). „ Undertake sensitization of communities and men in beneficiary households to reduce the perceived threat to men’s status as primary providers and thus reduce the risk of backlash. „ In consultation with beneficiaries, develop strategies to reduce the risk of opportunistic GBV, such as by organizing joint travel to program activities, locating work sites near beneficiary households, and so on (see Section 8 Provide). 40 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 8 PROVIDE 8.1 THE TRANSFER OF BENEFITS The benefit transfer is a core phase in the recurring implementation cycle and is often one of the main points of contact between a program and beneficiaries. Cash transfers are commonly used to smooth consumption and encourage investments in human capital development, while programs that intend benefits to be used for more specific purposes may choose to transfer in-kind goods or assets (such as food or agricultural inputs) or near cash (such as vouchers or fee waivers). The COVID-19 pandemic has accelerated a trend in many countries toward digitizing payments, enabling governments to deliver safe, secure, swift, and convenient payments at scale. This section focuses on how the design and operation of transfer methods can reduce program-related GBV risks and contribute to women’s empowerment and to broader GBV prevention. This section focuses primarily on cash transfers, but the design and implementation considerations are also generally applicable to transfers of in-kind and near cash benefits. While several factors related to program objectives and delivery systems come into play in choosing the transfer method, it is also important to consider whether women are likely to retain control over the transfers and whether conflict over the use of the transfers is likely to arise. In some instances, for example, if women are responsible for the purchase and preparation of food, they may be more likely to retain control over the transfers if the transfers are supplied as food or vouchers, which are less likely to be confiscated to be used for other purposes. However, this only makes sense if raising food security is a core development objective of the SSN program and if it is an efficient way to transfer resources in light of market conditions. In other cases, for example, if the objective of the transfer is to enable women to make human capital or livelihood investments, cash will be the preferred transfer method. It is important in this case to make design and implementation choices that increase the likelihood women will retain control over the cash and that reduce the risk of conflict or the confiscation of the transfers. This can be done by making smart design and implementation choices on the transfer of resources. Provide 41 EVIDENCE BOX 8. BENEFIT TYPE While the evidence is limited, studies comparing different transfer modalities to women find no differential impact on IPV. Results in Ecuador indicate that transfers lessen controlling behaviors and physical or sexual violence by 6–7 percentage points (or approximately 19 percent to 30 percent), but that the impacts do not vary by transfer modality (food, cash, or voucher).a There is suggestive evidence that a reduction in IPV was caused by improved women’s bargaining power, increased domestic labor of both husband and wife, and decreased poverty-related stress, while the consistency across modalities was attributed to the framing of the transfers as part of a wider food security intervention that did not aim to challenge traditional gender roles. In Bangladesh, cash and food transfers were similar in their lack of impact on IPV 6–10 months after the program ended unless they had been accompanied by behavior change communication.b While transferring either cash or food alone had no impact on emotional or physical IPV, transfers of either, combined with behavior change communication, led to a reduction in physical violence by 7 and 6 percentage points for cash and food transfers, respectively. a. Hidrobo, Peterman, and Heise (2016). b. Roy et al. (2019). 8.1.1 Payment systems There has been an evolution in SSN payment systems from manual toward digital methods, and payment systems fit along a continuum from manual to automated payment administration to digital payment provision.78 Some SSN programs operate separate systems in different regions, depending on the financial infrastructure. The guidance in this subsection focuses on the last mile of transfer to the recipient (and not the steps that occur earlier in the transfers from the program to payment agents), that is, whether the final transfer is digital or is transferred in-person in cash, in near-cash, or in kind.79 Many supply- and demand-side considerations go into the design of payment systems, including cost, capacity constraints, coverage and accessibility of payment operators (banks, post offices, and so on), mobile phone ownership rates and network coverage, and access to internet banking, and GBV considerations can help inform the final choice and design. KEY QUESTIONS „ For each transfer method under consideration, what is the likelihood that women will retain control over the resources transferred (and of the means of accessing the transfer, such as a debit card or a SIM card)? Are there risks that the transfer (and means of accessing the transfer) will be captured by other household members? „ Do women possess the identity documents required to enroll in the different modalities under consideration, in particular digital payment systems if they have specific requirements? If not, can the program facilitate access to 78 Automated payment administration is a system whereby back-end payment processes are automated, but payments may still be provided in-person. In digital payment provision systems, payments are made virtually through bank transfers, mobile wallets, postal services, and so on. See Lindert et al. (2020). 79 Transfers that must be cashed out at the points of transfer (that is, the transfers cannot be retained in a virtual form) present similar advan- tages and disadvantages to manual transfers in triggering, reducing, or preventing GBV. 42 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE identity documents? Or can the program obtain waivers or alternative means of establishing identities of the beneficiaries without legal proof of identity?80 „ Are there legal barriers to women’s independent control of financial resources, for example, a requirement for the husband’s consent before a woman can open a bank account? „ If women are required to travel to payment points, how can any risks of opportunistic harassment or assault be reduced? „ What measures can be established to prevent or mitigate SEA risks during interactions with payment operators or agents? „ Could interaction with program actors at payment points or payment system technology be leveraged to disseminate information to women, including on access to GBV support services? „ Can digital transfers be leveraged to close other gender gaps (in IDs, mobile phone ownership, bank account ownership, or digital and financial literacy)? Digital transfers paid to women offer opportunities for women’s empowerment and GBV prevention if they are accompanied by measures to address women’s constraints on access. There are four significant advantages of digital payments as a means to reduce GBV risks and empower women: (a) increasing financial inclusion and the ability to retain control over transfers or conceal them from violent partners, (b) lessening the need to travel to transfer points (thus reducing time burden, travel costs, and exposure to the risk of opportunistic harassment or assault during travel to transfer points), (c) lowering the risk of SEA by program actors, and (d) using dedicated communication channels for the dissemination of information on the SSN program or services, including available GBV support services, if the digital payment occurs through mobile phones that women own or control or if messages can be added to ATM receipts. However, manual cash payments may be preferable in areas with low coverage by payment service providers or if regular contact with program staff is desirable. Manual cash payments are often the only option because of the lack of digital payment services. Cash payments also have other advantages including (a) maintaining regular face-to-face contact between program staff and beneficiaries, thus creating the opportunity to deliver training and other accompanying measures that can contribute to GBV prevention, as well as gathering beneficiary feedback and potentially troubleshooting issues immediately; (b) providing easier access to women who have limited financial literacy or who do not possess legal IDs; and (c) preventing women’s reliance on an intermediary if the access to necessary technology is controlled or can be confiscated by household members (for example, if the rate of women’s individual mobile phone ownership is low or if single phones in households are controlled by a male household member). Depending on the context, the payment days may also present a unique opportunity for women to leave the home and interact with other women, thus building social capital. Evidence Box 9 presents emerging evidence on this issue. 80 See AFI (2019) for innovations to increase access to financial services. Provide 43 EVIDENCE BOX 9. MANUAL VS. DIGITAL PAYMENT METHODS The quantitative research comparing the impacts of manual and digital payments on the prevalence of GBV is limited, but existing research suggests that digital payments may be less burdensome and more likely to allow women to maintain control over resources.a A study in Niger randomly assigned women to receive transfers through either mobile payments or conventional cash transfers provided physically at central locations. The study did not consider impacts on GBV, but it did find that, relative to women who received conventional transfers, mobile money recipients spent less time traveling to payment points and waiting for the transfers, were generally more likely to be engaged in productive economic activities, and spent more on children.b Researchers hypothesize that women’s enhanced ability to conceal mobile money transfers boosted their ability to align expenditures with preferences compared with women in the physical cash group. This accords with other research suggesting that women recipients are willing to pay to maintain control over transfers and conceal them from their spouses.c Although these studies focus on the ability to conceal as a pathway to the better use of resources, concealing transfer amounts may, in some cases, also help prevent men’s backlash and men’s opportunities for extracting resources through violence. A study of Mexico’s Prospera SSN found that, if beneficiaries received debit cards allowing them to access their transfers at any ATM or to make purchases, the median road distance to access the accounts was reduced from 4.8 to 1.3 kilometers, thereby reducing the risk of opportunistic GBV.d Transferring money directly to women does not always enhance economic empowerment or autonomy; however, emerging evidence points to promising design features, as follows. „ Transferring funds through direct deposits to women’s accounts or mobile payments improves women’s control over the use of financial resources. In India, depositing wage payments for a federal workfare program directly to women’s bank accounts and training women on the benefits of the accounts raised the likelihood that women worked.e This seemed to enhance their bargaining power and led women to hold more liberal attitudes toward women’s work and mobility. Likewise, in Uganda, microfinance loans to women borrowers distributed through mobile money, rather than in cash, boosted business profits by 15 percent; the greatest impacts occurred among women who had experienced pressure to share money with their spouses.f „ Women’s savings accounts that offer commitment devices or are costly to access protected women’s funds against the demands of others. In the Philippines, access to a commitment savings account, which restricted withdrawals until a specified date or balance had been reached, increased women’s power over household decisions.g Similarly, if they were offered free bank accounts with no interest, but large withdrawal fees, 40 percent of women market vendors in rural Kenya used their accounts and exhibited a nearly sixfold increase in daily savings.h The relatively high withdrawal fees, which served as a commitment to avoid spending the funds saved, may have helped women protect their savings against pressures to share resources with family or friends. In Kenya, offering ATM cards that increased the accessibility and reduced the costs of using savings accounts led women with low levels of decision-making power relative to their spouses to stop using the savings accounts.i a. See Garz et al. (2020) for an overview of evidence on the impacts of digital payments. b. Aker et al. (2016). c. Almås et al. (2018). d. Bachas et al. (2018). e. Field et al. (2019). f. Riley (2020). g. Ashraf, Karlan, and Yin (2010). h. Dupas and Robinson (2013). i. Schaner (2017). 44 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE If payments are provided digitally, different accompanying measures can be a means for building social capital. The trade-off between the advantages of digital payments and the opportunities for women to come together during payment days will depend on the country context. However, other accompanying measures can be delivered to women receiving digital payments such as training on digital and financial literacy that not only tackles gender gaps, but also enables women to strengthen their social networks, which, evidence suggests, may help reduce IPV. Mobile phone networks and ATM receipts also provide a system for disseminating information to program beneficiaries, including on available GBV support services. For example, in programs relying on mobile networks to make transfers, information can be distributed through automated calls or text messages (although care must be taken to ensure that women can retain sole control of their phones before sending sensitive information). Access to mobile phones also supplies opportunities for women to seek information on obtaining services. This is particularly helpful if other forms of direct communication, such as during health crises (COVID-19 pandemic, for instance), are required.81 Another advantage of using digital networks as a point of contact with transfer recipients is the opportunity the contact creates for establishing an entry point for confidential complaints through a program’s grievance mechanism (see Section 9.2 Grievance mechanism). Similarly, in programs through which payments are made using bank transfers, ATM receipts can include information on access to GBV support services or other messages to raise awareness related to GBV and the avenues available to report cases. If a program provides digital payments, giving beneficiaries the choice between multiple service providers has several advantages for program efficiency and GBV prevention. Experience in Zambia shows that this can benefit both the program and the beneficiary (Good Practice Box 10). It increases competition, reduces costs, and provides fallback options if any provider fails to offer quality services. For the beneficiaries, increased competition tends to drive down transaction fees and create incentives for providers to offer additional financial services as they compete for customers. Users may also more easily choose or switch to the most convenient and safest transfer modality that allows them to retain control over the transfer (for example, choosing among mobile transfers, bank transfers, post offices, or other touch points). If women’s mobile phone ownership rate is low, the program could consider providing SIM cards or low-cost mobile phones to women or incentivizing providers to do so. This asset transfer could also boost women’s control and agency. Programs may choose to continue in-person cash transfers in remote areas or if digital connectivity rates are low, while providing digital options if there is more access. 81 The frequency of calls or reports has risen by more than 25 percent in Argentina, Cyprus, France, and Singapore during the pandemic. Provide 45 GOOD PRACTICE BOX 10. A CHOICE-BASED PAYMENTS SYSTEM IN ZAMBIA The Girls’ Education and Women’s Empowerment and Livelihoods Project (GEWEL) in Zambia has developed an innovative, multi-provider, digital payment system that is centered on beneficiary choice. In a departure from traditional cash transfers delivered through civil servants, the system promotes the financial inclusion, agency, and self-determination of women beneficiaries. Participants in GEWEL’s economic inclusion component can decide for themselves the provider and the type of account they want to use for grants (commercial bank accounts, mobile wallets, ATM cards, or post office accounts). Various elements have been critical to the model’s success in contributing to women’s empowerment, as follows: „ Choice of providers: During enrollment, recipients choose the provider and account where they prefer to receive payments. This promotes women’s agency and financial inclusion. „ Financial literacy training: Accompanying measures—business and life skills training—were expanded to provide more detailed, yet easy-to-understand information on payment services and providers. This is particularly important in a choice-based payment model to enable beneficiaries to make informed decisions. „ More money to the recipients: Transfer recipients receive a top-up to cover withdrawal fees. The top-up is pegged to the most expensive service fee on the market, allowing recipients to pocket the difference if they choose a provider with lower fees, thus increasing competition between providers and driving down the fees charged to clients. This is a novel approach compared with the traditional system whereby governments pay providers a fee to bring payment points closer to recipients and to remove cash-out fees for recipients. „ Identification of best providers by community volunteers: Community-based volunteers were paid through the choice-based payments system not only to facilitate the payment of stipends, but also to incentivize the volunteers to identify the best local provider options available and offer more informed guidance to recipients. The decision tree in Figure 6 presents the opportunities and trade-offs that are associated with payment method design choices. 46 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE FIGURE 6. Payment method design choices Payment modality In-person payment Digital, mobile cash In cash, in near-cash, or in kind Opportunities Risks Opportunities Risks • Contact between program • Travel-related risks • Increase financial inclusion • Di culty for women with actors and beneficiaries • SEA risks (by payment • Increase control over the use limited digital and financial provides opportunities to agents) of resources literacy to understand and deliver accompanying • Cash may be more easily • Increased confidentiality of access virtual cash measures at the same extracted through the use or transactions • SEA (by payment agents) moment the threat of violence • Opportunity to build digital • Potentially longer distances • Accessible to women with • Di cult access for the and financial literacy to pay points; need to take limited digital and financial elderly, persons with public transportation, which literacy disability, remote groups, may expose women to GBV • Inclusive of beneficiaries and so on without ID cards and those living in remote locations Mitigation Mitigation • Assistance in setting up • Designate safe and accessible bank or postal accounts or transfer points and facilitate mobile wallets safe transportation • Financial literacy training • Strict oversight of manual transfers DESIGN AND IMPLEMENTATION TIPS All methods „ Identify the type of transfer (cash, near cash, or in-kind) that is more likely to remain under the control of the designated recipient and to be used for the intended purposes. „ Determine the accessibility constraints on women, the elderly, persons with disability, or other groups of beneficiaries in reaching transfer or payment points and make accommodations as needed. „ Establish criteria to identify alternate recipients (if designated recipients become unable to collect the benefits themselves) and monitor the use of the alternates to prevent the capture or misuse of transfers and to avoid the confiscation of beneficiary cards or other required identity documentation. „ Take measures to ensure accessibility, including among people with disabilities or the elderly, and institute additional safeguards or identify alternative payment mechanisms for target groups if necessary. „ If cash transfers are chosen and if access to the appropriate technology is available or can be provided by the program, shift toward digital payments, which tend to be more secure and increase women’s financial inclusion. Provide 47 Digital cash payments „ Confirm that the requirement to present proof of identity to receive payments does not place an undue burden on women. If there are constraints, identify alternate forms of ID or provide assistance in obtaining the necessary identification documents.82 „ Ensure that all recipients have access to the required technology (debit cards, ATM machines, mobile phones, and so on) and, if possible, give them the choice of payment provider so they may choose based on accessibility and convenience, as well as the provider’s terms and conditions. „ If payments are transferred to women, monitor women’s ability to retain control over the necessary technology (debit cards, mobile phones, and so on) or increase their control through provision of the technology to them. „ Provide digital and financial literacy training and any necessary assistance in the use of technology. Complementary measures to improve women’s financial literacy not only enhance women’s ability to manage and retain control over resources, but can also build social capital, strengthening GBV prevention. „ Identify alternative payment mechanisms for any beneficiary groups that face specific access constraints. „ Identify any risks of SEA by payment providers; establish mitigation measures (including the dissemination of information on procedures for reporting abuse through the program’s grievance mechanism); and ensure regular monitoring is in place. „ Assess the SEA/SH codes of conduct and accountability mechanisms of service providers and train them if necessary and feasible. „ Establish strong grievance redress channels independent of the payment service providers, whereby beneficiaries can submit complaints about any difficulties or abuses of power they experience in seeking to access payments. „ Prepare to reconsider delivery mechanisms and modalities if these exacerbate or create protection risks or if there are reports of abuse. In-person transfers of cash, near-cash, or in-kind benefits (in addition to the above) „ Choose secure and accessible locations and schedules for manual transfers. Consider coordinating with local authorities to increase the safety at transfer points. „ Take steps to reduce the security risks faced by women traveling to or from transfer points, for example: ƒ Establish transfer points in safe locations as near as possible to beneficiary households ƒ As much as possible, conduct transfers at times that do not require beneficiaries to travel outside of daylight hours; take the domestic responsibilities of recipients into account ƒ Encourage beneficiaries to travel in groups to transfer points ƒ Avoid public disclosure of transfer days, times, and locations 82 See World Bank (2018) for guidance on digital onboarding to access digital payment systems. 48 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE „ Ensure robust oversight of in-person transfers to reduce the risk of SEA and inform beneficiaries of channels through which they can report any grievances or abuse. „ Identify and mitigate any increased risk of SEA by transfer agents in FCV contexts or during rapid scale-up in response to shocks. In FCV contexts and during shocks, beneficiaries are particularly vulnerable to exploitation and abuse given their heightened dependence on transfers. 8.1.2 Transfer size, frequency, and duration There are many considerations in design choices that revolve around the size and frequency of transfers, such as the objective of the program, the costs, and the trade-offs between program coverage and the benefit level. Generally, programs provide larger, lumpier transfers if they seek to allow households to purchase assets and invest in productive activities, whereas smaller, more frequent transfers are used to smooth consumption and make human capital investments. In deciding on transfer amounts and frequency, programs also often take into account the expenses incurred by beneficiaries in traveling to payment points, including the opportunity cost of women’s time spent collecting transfers. This subsection considers the extent to which the size, frequency, and duration of transfers contributes to women’s empowerment, GBV prevention, and the mitigation of any program-related GBV risks. KEY QUESTIONS „ What are the trade-offs between the size and frequency of transfers relative to women’s ability to retain control over program resources? „ What is the size and frequency of transfers that are most likely to lead to men’s backlash? How can the risks be reduced? Are there cultural or contextual factors that may exacerbate these risks? Provide 49 It is often thought that smaller, regular transfers (conducive to small household purchases already managed by women) are less likely than larger transfers to be seen as a threat to the role of men as the primary breadwinner.83 However, the evidence is mixed (Evidence Box 11). EVIDENCE BOX 11. THE SIZE AND FREQUENCY OF TRANSFERS The evidence is inconclusive as to whether smaller, more frequent, or lumpier transfers are better for women’s empowerment and the prevention of GBV. Initial research in Mexico suggested that large payments, unlike small payments, are associated with increased violence.a This was thought to show that there was less incentive for men to use violence to extract smaller amounts than larger sums. However, a study in Kenya testing lump-sum versus periodic transfers reported a significant improvement in a women’s empowerment index following a lump-sum transfer.b The index incorporates measures of the frequency of physical, sexual, and emotional abuse by husbands and the justifications offered for violence against women. Some qualitative evidence suggests that, with smaller transfer values, men are less threatened in their role as primary providers, and a backlash therefore becomes less likely.c Smaller transfers may also be more easily hidden by women and make extracting transfers from women less of an incentive. However, a study in northern Nigeria that varied the frequency of transfers did not find any difference in women’s control over resources.d The regularity and predictability of transfers are also important in reducing GBV risk. Research on a conditional cash transfer program in Colombia demonstrated that unanticipated missed payments are associated with an increase in violence because the missed payments create an adverse emotional response.e However, IPV rates declined by approximately 5 percent around the time of receipt of the transfer. a. Angelucci (2008). b. Haushofer and Shapiro (2018). c. CaLP (2018). d. Bastian, Goldstein, and Papineni (2017). e. Camacho, Gaviria, and Rodríguez (2016). 83 Wasilkowska (2012). 50 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Figure 7 presents a decision tree of selected gender- and GBV-related issues and trade-offs to consider in making decisions about the size and frequency of transfers. FIGURE 7. Design choices: transfer size and frequency What is the optimum size and frequency of transfers? Smaller and more frequent Larger and less frequent Opportunities Risks Opportunities Risks • Women more likely to retain • Insu cient amount for • Opportunity to make • Greater incentive to use control over resources and economic empowerment or productive investments that violence or threat of less likely to trigger backlash shifts in bargaining power can sustainably improve violence to extract resources livelihoods and increase • Potentially more threatening economic independence to men as primary breadwinners, leading to risk of backlash Mitigation • Accompanying measures to support women's economic empowerment" because the Mitigation mitigation measure is not related to transfer size/fre- • Communication about the quency (or linked to the risks) purpose of the transfer but payment modality • Consider involving men and (previous section) boys in program activities to secure buy-in for the objectives To the extent that the duration of transfers (program exposure) affects the sustainability of poverty reduction, it is thought to be an important determinant of the sustainability of any program-induced reductions in violence within beneficiary households, regardless of the sex of the transfer recipient. If a household falls back into poverty once benefits have ceased, poverty-related stress would be expected to return, likely reversing any gains in violence reduction (Evidence Box 12).There are other mechanisms through which programs can contribute to GBV reduction, such as women’s empowerment and shifts in gender norms, which, if successful, can strengthen the sustainability of any reduction in GBV, even if poverty-related stress reemerges. Provide 51 EVIDENCE BOX 12. THE DURATION OF TRANSFERS A short-term emergency cash transfer in Malawi was targeted at men heads of household who were selected for their traditional provider role. Many men described how receiving the transfers allowed them to be viewed positively because they were fulfilling their male role as provider, and there were fewer conflicts over limited resources. However, this improved status was temporary. “They look at us as providers… Man is considered a provider. If you get CT [a cash transfer], you are providing. Everyone looks at you and is happy. But now it [the cash transfer program] has ended; quarrels may start.”a However, transferring resources to men increases their control over household resources and therefore women’s dependence. Gender inequality, including in access to resources, is an underlying driver of IPV and other forms of GBV. Transferring resources to women is therefore generally preferable, unless gender norms are so rigid that the risk of backlash becomes significant. a. CaLP (2018, 27). DESIGN AND IMPLEMENTATION TIPS „ Take steps to increase the likelihood that women will retain control over transfers; this may be accomplished through ongoing messaging on the intended use of program resources, social contracts, support for household budgeting, monitoring, and behavior change communication and activities. „ Prioritize the predictability and reliability of transfers; communicate clearly in advance any delays or pauses in the transfer schedule. „ If feasible, choose a payment system that allows women to decide whether they want to withdraw small amounts multiple times (without incurring transaction fees), rather than cashing out all benefits at once. „ Ensure that beneficiaries and their partners understand clearly the program exiting process to avoid unexpected losses of income that could trigger intrahousehold conflict. 8.2 ACCOMPANYING MEASURES Many cash or in-kind transfer programs include accompanying measures or complementary activities, often referred to as cash plus. While the content, frequency, and duration of these activities vary, they are usually delivered in group settings and cover topics such as hygiene and feeding practices to improve nutrition outcomes or early childhood stimulation and care. Some schemes may include more extensive skills training. Graduation or economic inclusion programs thus usually involve training on livelihoods and business management. The focus of this subsection is group sessions. These are routinely delivered to safety net beneficiaries by facilitators or volunteers and typically key on conveying simple messages about behaviors to improve human capital outcomes. Programs are increasingly drawing on behavioral science to inform the design of the accompanying or cash plus measures. 52 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE If cash transfers are combined with accompanying measures, they are more likely to reduce IPV than if the transfers or the measures are delivered separately, even if GBV prevention is not an explicit objective of the activity. Evidence from Bangladesh demonstrates that group sessions designed to improve nutrition outcomes strengthened women’s social capital, leading to a reduction in IPV (Box 15).84 It seems that there are a number of positive impacts associated with simply participating in group activities that contribute to the reduction of IPV. Participating in group activities can strengthen women’s support networks. It can also strengthen women’s social status as knowledge and skills bearers and, in so doing, raise the social sanction or cost to men of the use of violence, thus serving as a deterrent on IPV. This subsection examines design features that may maximize the potential of accompanying measures to promote empowerment as well as design and implementation features that may minimize the risks associated with accompanying measures (whether the risks related to GBV or not). It also provides a summary of the recent evidence on interventions that are designed explicitly to prevent GBV and that could accompany cash transfers if appropriate. KEY QUESTIONS „ Are there entry points for accompanying measures to nudge social norms toward increased gender equality and the rejection and delegitimization of GBV? Are there opportunities for engaging men and boys in sessions on masculinity and more equal gender roles, including more equal sharing of care work? Can broader support in the community be built for more equal gender roles? „ Are accompanying measures delivered in ways that are safe and convenient for women? Under what conditions and in relation to which issues should accompanying measures be delivered in women-only safe spaces without the presence of men? „ Are accompanying measures delivered to be conducive to strengthening women’s networks, social capital, knowledge, and agency? „ Are there existing women’s groups or organizations with experience working on gender or GBV issues that can inform the design and delivery of accompanying measures? „ What are the opportunities and challenges in the effort to increase the number of women interacting with beneficiaries who are engaged in accompanying measures through the recruitment and retention of front-line service providers? „ Are there any hierarchies in households and communities that should be taken into account in designing accompanying measures? „ Does travel to and participation in accompanying measures create risks of SEA or opportunistic harassment or assault that need to be mitigated? 84 Roy et al. (2019). Provide 53 EVIDENCE BOX 13. STRENGTHENING WOMEN’S SOCIAL CAPITAL Group-based activities build social capital and networks that can increase women’s confidence and social status, raise the costs of the use of violence by men, and, in cases of GBV, either help resolve the conflict or provide support to women seeking to exit abusive relationships.a This appears to be valid even if issues of gender or GBV are not explicitly addressed.b In a randomized control trial of a cash plus program in Bangladesh, sustained reductions in IPV were achieved only if transfers were combined with nutrition training sessions.c The sustained impacts were attributed to the increased social capital and reduced tolerance for IPV among beneficiaries, the increased social cost to men of the use of violence, and the strengthened support network in cases of IPV. Viewing accompanying measures as an opportunity to create safe spaces or women’s clubs has also been found to help establish social networks and provide relief from isolation and boredom in contexts such as refugee camps where women may worry for their physical safety and find the only solution is to remain in isolation.d Research has shown that the relationships fostered in the spaces created by women’s rights organizations and networks often lead to changes that are associated with empowerment.e The changes include a growth in self-confidence, new capabilities, greater awareness of gender inequalities, and the capacity to act collectively to demand the recognition of rights. Participation in group activities reduces isolation (which is often an element of GBV and controlling behaviors by intimate partners) and provides links to informal support networks for beneficiaries experiencing GBV. Enhancing women’s social networks could thus be a powerful force for reducing IPV, a finding that extends beyond SSN interventions. Accompanying measures are more likely to have an IPV reduction impact if they are designed to build women’s self-esteem, effectiveness, and communication skills. Some participants in a South African program, IMAGE, reported that the increased self-confidence, social support, and communication skills gained from participation in a combined microfinance and training initiative resulted in improved partner communication that helped prevent conflicts from escalating into violence.f Similarly, interventions that build up aspirations as part their plus activities have also been found to be effective in reducing IPV risk. Cash transfers in Kenya reduced IPV if they were accompanied by the screening of a 20-minute video showing positive role models.g If they were delivered alone, neither the cash transfers nor the aspirational video had an impact on IPV.h a. Brody et al. (2015); Stets (1991). b. Roy et al. (2019). c. IPV did not differ between women receiving transfers and a control group 6 to 10 months after the program. However, women who received transfers, along with behavior change communication, experienced 26 percent less physical violence (Roy et al. 2019). d. CaLP (2018). e. Pathways of Women’s Empowerment (2011). f. Kim et al. (2007). g. Mahmud, Orkin, and Riley (2020). h. In northern Nigeria, reductions in IPV were sustained only if cash transfers programs included add-on measures linked with community-wide livelihoods support (Cullen, Gonzales Martinez, and Papineni 2020). Accompanying measures are now tackling gender norms (sometimes including GBV) more explicitly. Research is ongoing to identify the most effective approaches to achieve a shift in gender norms: working with women beneficiaries exclusively, engaging men, working with couples, engaging with traditional leaders, or undertaking dialogue in communities more broadly (Box 14). The best approach will be context specific and depend on social and cultural norms, including the extent to which discussion of gender or GBV is taboo and risks triggering backlash and the extent to which there is recognition that these issues can be addressed openly. In places in which gender issues and GBV are sensitive or considered taboo, it may be more effective to nudge the gender norms into the open for discussion through activities that address the underlying drivers of violence, rather than confronting GBV directly. It may be possible to 54 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE balance gender roles without making this objective explicit, for example through accompanying measures that focus on more participatory household decision-making, socioemotional health, substance abuse, or conflict resolution within households. EVIDENCE BOX 14. SHIFTING GENDER NORMS Accompanying measures are increasingly engaging husbands and men community members in the effort to change social norms and build support for women’s economic empowerment as a way to reduce the risk of GBV. This has occurred largely because of the recognition that programs targeting women need to reflect the complex dynamics of the relationships of beneficiaries with men rather than regarding the beneficiaries as autonomous agents. Even if men are not the direct program participants or recipients, they often exert substantial influence on women’s participation in program activities or the way in which transfers are used. Moreover, the interactions between men and women are governed by complex networks of social norms. Altering gender roles thus requires the engagement of men. Several initiatives have succeeded in improving the outcomes among women by engaging men, including the United Nations Population Fund’s schools for husbands in Niger, CARE’s couples training to build support for women’s participation in savings groups, and Promundo’s participatory couples discussions on topics such as gender, power, and masculinity. Some of the most promising evidence on these intensive gender-focused groups and workshops is supplied through couples training programs and Promundo’s early parenting program.a Others with promise include SASA! in Uganda, Steppingstones, Journeys of Transformation, and Sonke Gender Justice.b Two studies that evaluated the impact of the International Rescue Committee’s Engaging Men through Accountable Practice Program focused primarily on the effort to change men’s attitudes in conflict-affected settings, such as in Côte d’Ivoire and the Democratic Republic of Congo, but did not find statistically significant reductions in IPV. It thus appears that programs working together with couples may have more success.c The low cost and ease of scalability of educational entertainment interventions (edutainment) have generated growing interest. In India, the arrival of cable television reduced the acceptability of GBV.d Although public video screenings in Nigeria and Uganda have not led to statistically significant reductions in IPV, they have changed attitudes toward risky sexual behaviors.e Cheaper than television and print, radio may also be particularly effective at reaching rural populations, as well as populations in fragile, conflict, or postconflict settings.f Radio programs that were not specifically designed to seek to change attitudes or behavior, but which portrayed empowered women and women having small families, have been shown to impact fertility and women’s decision-making power.g Targeted radio and television edutainment interventions have caused substantial changes in perceptions around violence, helped enhance awareness, and boosted support for legislation addressing domestic violence.h However, more evidence is needed on why edutainment appears to affect attitudes, but not behaviors associated with IPV in some settings and behaviors, but not attitudes in others.  ee Doyle et al. (2018). Promundo introduced a program in Brazil as a companion to the Bolsa Família cash transfer to try to change gender norms and a. S mitigate any risks of violence. See “Bolsa Família Companion Program,” Promundo, Washington, DC, https://promundoglobal.org/programs/bolsa-famil- ia-companion-program/. b. On SASA!, see Abramsky et al. (2014). See Hossain et al. (2014); Jewkes et al. (2020); Vaillant et al. (2020); World Bank, GWI, and IDB (2014). The following websites provide rigorous evidence on c.  what works in preventing GBV: Promundo, at https://promundoglobal.org/; SVRI (Sexual Violence Research Initiative), at http://www.svri.org/; What Works to Prevent Violence, at https://www.whatworks.co.za/. d. Jensen and Oster (2009). e. Banerjee, La Ferrara, and Orozco (2019); Green, Wilke, and Cooper (2020). f. Armand, Atwell, and Gomes (2020); Bilali, Vollhardt, and Rarick (2016). g. Cheung (2012); Jensen and Oster (2009); La Ferrara, Chong, and Duryea (2012). h. Usdin et al. (2005); Yue, Wang, and Singhal (2019). Provide 55 Little is known overall about the sustainability of the impacts of interventions focused on gender relations or IPV prevention. Researchers hypothesize that reductions in the prevalence of IPV resulting from women’s empowerment may be sustained to the extent that empowerment is sustained, that is, if a rebalancing of bargaining power, improvements in marital relations and collaboration, and reductions in women’s dependence on their intimate partners are sustained. Similarly, lower levels of social acceptability of GBV and increased costs to abusers would only be expected to persist if the shift in attitudes toward GBV is sustained. Creating conditions conducive to open discussion and frank communication is important for the success of any accompanying measures. This requires an understanding of specific social dynamics and social hierarchies where activities are conducted, and thus it is important that these be identified as part of the risk assessment process (see Section 5 The environmental and social framework). In some settings, it may not be socially acceptable for women to speak openly in front of men, particularly on sensitive issues such as gender norms and GBV. In such cases, it may be necessary to create safe spaces for women that are conducive to open discussion. There may be other hierarchies within households and communities that should be taken into account in designing accompanying measures. For example, in some societies, wives may not speak openly in front of their mothers-in-law and younger wives in polygamous households may not feel free to speak in front of more senior wives, while, in other settings, there may be tensions among different ethnolinguistic groups. In such cases, separate activities may be more effective. Even if accompanying measures do not explicitly address GBV issues, engaging women in group activities provides an opportunity to disseminate information on available GBV support services. Most accompanying measures in SSN programs do not primarily aim at GBV reduction, but there is a growing body of evidence on other interventions that are explicitly designed to reduce violence. The What Works to Prevent Violence against Women and Girls Global Program has evaluated 15 interventions designed to reduce violence against women and girls.85 The emphasis is on addressing physical and sexual violence by intimate partners, violence in the family, and bullying and violence in schools in Central and South Asia and in Sub-Saharan Africa. Four types of programs for violence prevention were examined. The first two—women’s empowerment approaches and couples interventions—are adaptable and are included among accompanying measures in SSN programs. The remaining two—community activism approaches and school-based violence prevention among children—may fall outside the scope of most SSNs (Evidence Box 15). 85 Jewkes et al. (2020). 56 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE EVIDENCE BOX 15. THE WHAT WORKS TO PREVENT VIOLENCE PROGRAM Economic empowerment approaches Five interventions that combined women’s empowerment with an attempt to shift gender norms were evaluated. The interventions—in Afghanistan, Bangladesh, Nepal, South Africa, and Tajikistan—were assessed for their effectiveness in preventing physical or sexual IPV. Key design features of successful interventions that were identified included providing sufficient funds that remained under women’s control; focusing on IPV based on a contextual understanding of the drivers of violence; working with women, men, and extended families; reliance on highly selected, trained, and supported personnel; and sessions that were sufficiently long (two–three hours) and frequent (weekly) and that were delivered over an extended period (four–six months). The results of the evaluation suggests that focusing only on women may be insufficient to change gender relationships fundamentally in highly patriarchal societies in which the power of young women is particularly constrained. In these contexts, interventions with women, men, and entire families may be more effective. All interventions that involved men were effective in changing men’s behavior. The inclusion of an economic component was attractive to men, leading to higher retention rates and enabling a conversation with men about gender, social norms, and power dynamics. In Nepal and Tajikistan, working with multigenerational families had positive impacts on households and strengthened women’s economic position in the household. Qualitative research in both countries indicated that family-centered models can be effective in enabling women to participate fully without being viewed with suspicion and that this can help build trust and tamp down the risk of violent backlash (none was reported). Couples interventions What Works evaluated four interventions among couples, all involving counseling and group-based curricula in India, Nepal, Rwanda, and Zambia. The primary objective of all four interventions was to prevent physical or sexual IPV. The main outcome was assessed two or more years after the baseline, except in Zambia, where it was measured after one year. The four evaluations were all randomized controlled trials. The evaluations identified specific elements of intervention design and implementation that determined the degree of each program’s success in reducing IPV. These included well-designed theories of change adapted to the specific context; inclusion of only well-established couples as program participants; the testing of newly developed interventions or adaptations before implementation; sufficient program exposure (40–50 hours) over an extended period (weekly for four–six months); the delivery of interventions by experienced facilitators; the pilot implementation of the curriculum before the start of delivery; and careful training prior to the start of the intervention, followed by supervision and support among facilitators who had not previously delivered the interventions. Community activism approaches These interventions trained community members as volunteer activists. The more successful interventions worked with action groups established within the community, rather than engaging community members as individuals. They used participatory methods in workshops or other activities to enable critical reflection on gender relations, the individual participant’s experiences (encompassing, among the men, their use of violence), skills building, and experiential learning. They developed materials and manuals to support implementation by all actors, including the community action team members. All successful interventions involved engagement with women or couples that had experienced violence and provided support for survivors. The more successful interventions had a large body of (mainly volunteer) staff and activities spanned a minimum of 18 months. Effective interventions carefully selected volunteers or received nominations from communities; personnel were known to possess the desired attitudes and model the appropriate behaviors before they had been trained to deliver the program. The more successful interventions generally provided longer training (two to three weeks), although this was not the case everywhere. Constant support for personnel was also a notable feature. Provide 57 The prevention of violence against children Four of the What Works evaluations focused on interventions to prevent violence experienced or perpetrated by children. Three of the interventions were school-based, and one was home-based. Successful interventions reflected social empowerment theory, applied gender-power analysis, and sought to build gender equity and foster positive interpersonal relations. They emphasized empowerment, critical reflection, communication, and conflict resolution skills. Their learning components extended over a longer period, integrated experiential learning, and supported change in behavior and ideas through an engaging pedagogy among children. The interventions also addressed multiple drivers of violence, including gender inequity and the role of social norms in the use of violence, as well as activities to build social and emotional skills. All interventions provided implementers with manuals and training materials, and the trainers were all carefully selected, trained, and supported throughout the duration of the interventions. Source: Elaborated based on Jewkes et al. 2020. DESIGN AND IMPLEMENTATION TIPS „ Encourage activities that strengthen social capital and support networks among women. „ Unpack how social norms may inform the design of accompanying measures, particularly those that aim to change gender relations. „ Test messages, consider piloting interventions before rollout, and monitor for unintended impacts, particularly if GBV is explicitly addressed in the intervention. „ Find opportunities to nudge gender norms in the delivery of accompanying measures, even if women’s empowerment or GBV prevention is not the explicit focus of the measures. „ Engage local leaders and norm influencers (religious and traditional leaders, elders, and so on) in support of accompanying measures and to prevent backlash against the perceived threat of women’s empowerment or shifting gender norms. „ GBV survivors should never be singled out for participation in accompanying measures because the risk of stigmatization and traumatization is significant. However, facilitators should receive guidance on how to respond if beneficiaries disclose experiences of GBV, including providing information on available GBV support services and on the program’s grievance mechanism (see Section 9.2 Grievance mechanisms). „ Decide on the locations and times of accompanying measures while taking account of women’s care work and the safety of the locations and of any required travel. „ To enable women’s participation and improve children’s well-being, consider whether childcare might be provided. „ Ensure that ethical research standards are respected during the collection of GBV data and in the evaluations of accompanying measures.86 „ Facilitate the recruitment of women front-line service providers. Ensure SH mitigation measures are established among front-line providers (for example, supplying safe transport). „ In contracts with service providers, include SEA/SH considerations, such as codes of conduct and training for providers. 86 Ellsberg and Heise (2005). 58 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 8.3 PROVISION OF SERVICES SSN programs, particularly those with human development objectives, often encourage or require beneficiaries to access health care or education services. Other SSN programs may facilitate women’s access to civil registration and financial or other services. These services are generally offered through other government programs and institutions that are beyond the mandate of most SSN programs. However, to the extent SSN programs call for beneficiaries to use these services and verify compliance, the mitigation of the risk of SEA associated with these processes also falls under the responsibility of the programs. In addition, discussions on institutional arrangements for the provision of complementary services represent an opportunity to create more awareness and strengthen the supply of support to SSN beneficiaries experiencing GBV. Similarly, discussions with World Bank colleagues on aligning GBV prevention activities across human development sectors are opportunities for a portfolio-wide approach to boosting national capacity for the identification, referral, and support of GBV survivors. KEY QUESTIONS „ Are there risks of opportunistic harassment or assault if beneficiaries travel to and from service delivery points? „ Are there risks of SEA if beneficiaries access services or if providers verify compliance with conditionalities? How can these risks be mitigated? „ Can the services to which SSN beneficiaries are referred or are encouraged to use be adapted to offer safe spaces for beneficiaries to report experiences of GBV? Are there opportunities to enable qualified providers to respond to any cases? Accessing services such as health care, education, and civil registration is essential for increasing women’s human capital and agency, which contribute to GBV prevention and empowerment over the longer term. However, depending on the context, accessing such services may challenge social norms. For example, accessing family planning may be taboo if men have traditionally taken such decisions. Thus, access to family planning is essential in building women’s human capital, but achieving a shift in the gender norms in family planning decision-making can lead to a backlash among men. Similarly, accessing civil registration and gaining legal identity are central to increasing women’s agency, some men may feel threatened by the resulting expansion in women’s independence. Engaging men and boys, as well as community leaders, is therefore an important strategy for reducing the risk of backlash. Local leaders can help build support for women’s access to basic services. Engaging men and boys during program outreach and in social mobilization in favor of the benefits that accrue to the entire household if women have access to basic services can be effective in reducing backlash. SSN programs that seek to enhance human capital development tend to include measures to incentivize service uptake. This can range from social mobilization and promotion to the inclusion of incentives and add-ons to basic benefits packages and to linking the eligibility for benefits contingent on using the services. Many considerations are involved in such decisions, for example, the quality and accessibility of services. The risk of SEA should also be identified and taken into consideration. If there are significant SEA risks associated with service provision (for instance, SEA by supervisors in girls dormitories at school) or the certification of compliance (such as the certification of school attendance or of vaccination), it may not be appropriate to condition SSN benefits on using the services. To the extent the services Provide 59 are promoted by the program, SSNs should work with the institutions responsible for the services to strengthen mitigation mechanisms, including codes of conduct and access to grievance mechanisms with the capacity to receive SEA complaints. Any risk of opportunistic harassment or assault of SSN beneficiaries during travel to and from service delivery points can be mitigated. Mobile campaigns promoting vaccination, civil registration, and other services can expand service accessibility and uptake, in addition to reducing travel-related risks. Encouraging beneficiaries to travel in groups or providing transportation in insecure areas is advisable. In addition, interactions with service providers can offer opportunities for beneficiaries to disclose experiences of GBV and receive support. Service delivery points can represent a safe space where women can disclose GBV, particularly if staff are trained to identify and respond to cases and support services are adequate. This is particularly relevant if SSN programs directly link beneficiaries with social and labor services or assign caseworkers who can refer beneficiaries to GBV support services. In programs in which community workers, program actors, or NGOs conduct home visits, these agents can be trained in the safe identification of at-risk women and children and in how to direct women properly to available GBV support services. Women experiencing GBV are more likely to report incidents to women service providers. Recruiting women as front-line providers is therefore important. GBV is most effectively tackled through a coordinated, multisectoral response (Section 4 Country engagement strategy). SSNs can provide an entry point for engaging counterparts in a broader dialogue on a national strategy for GBV prevention and the provision of support services to GBV survivors. Cross- sectoral efforts are critical to strengthening national capacities for the identification, referral, and support of GBV survivors. DESIGN AND IMPLEMENTATION TIPS „ If women are required to access services that challenge gender norms, such as family planning services, civil registration for women, and independent access to banking services, engage men and local leaders to build support for women’s access to these services. „ If the risk of SEA by service providers during the provision of services or during the verification of compliance is substantial or if mitigation measures are insufficient to lower the risk, consider relaxing conditionalities or changing verification processes. „ Ensure that codes of conduct are adequate and that a grievance mechanism for SEA/SH reporting has been established and is effective (see Section 9.1 Codes of Conduct and Section 9.2 Grievance mechanisms). „ If programs require beneficiaries to use services, ensure oversight of the verification process. „ Promote the use of automatic verification protocols and inform beneficiaries of opportunities to report abuses through the program’s grievance mechanism (see Section 9.2 Grievance mechanisms). „ Where relevant and possible, encourage the adaptation of service delivery points to provide safe spaces for reporting experiences of GBV and accessing support services. 60 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 8.4 PUBLIC WORKS AND ECONOMIC INCLUSION PROGRAMS Public works programs in developing countries are frequently associated with a dual objective: offering temporary or seasonal employment and building or maintaining local infrastructure or providing local services. In many cases, the programs are run alongside other SSN programs, such as the supply of income support to vulnerable populations through cash transfers. In addition to representing a source of income, participation in public works is also an opportunity to improve social status and self-esteem among vulnerable populations through the dignity of work. They have also been shown to reduce gender wage gaps and empower women.87 Public works that build gender-smart infrastructure can reduce women’s care burdens and the GBV risks associated with, for example, the lack of safe sanitary facilities and the distance to schools or water points requiring women and girls to undertake extensive travel to collect water or attend school. A few countries are experimenting with the provision of care services, such as childcare, through public works, thereby not only supplying an important service, but also increasing the recognition and compensation for care work. An economic inclusion program is a bundle of coordinated multidimensional interventions that support households and communities in their efforts to increase incomes and assets.88 Livelihoods grants or credits are sometimes implemented as part of a multidimensional social protection strategy that seeks not only to help the poor meet their immediate consumption needs through poverty-focused cash transfers or public works programs, but also to offer beneficiaries a pathway out of poverty by providing capital to invest in sustainable livelihoods. The same pathways by which cash transfers promote a downturn in IPV by reducing poverty and food insecurity, empowering women economically, and increasing women’s social capital and networks can be stimulated by livelihoods grants or credits supplied to women. The skills building and social interaction among beneficiaries—for example, through life and business skills training and savings groups—that are often integrated within livelihoods programs can strengthen women’s social networks and agency, thereby contributing to GBV reduction. Participation in livelihoods programs can be expected to have positive impacts similar to public works, that is, recognition of women’s work and expanding their control over income. However, as with public works, women’s participation in livelihoods programs may trigger GBV if these activities are perceived to threaten men’s role as providers or transgress social norms on women’s work or if they increase the exposure to opportunistic harassment or assault during the engagement with livelihood activities. As with all SSN programs, these risks can be reduced or mitigated through careful design and implementation. KEY QUESTIONS Questions relevant to all economic empowerment programs „ Is the type of work or livelihood activity supported by the program likely to challenge traditional gender norms? „ What opportunities are there to increase the acceptance of women’s engagement in income-generating activities or in employment outside the home? „ Are there specific types of public works programs that are more sensitive to women’s needs and constraints? 87 Subbarao et al. (2013). 88 See the website of the Partnership for Economic Inclusion, World Bank, at https://www.peiglobal.org/. Provide 61 „ What arrangements can be made to accommodate competing demands on women’s time because of women’s caregiving roles? Are there opportunities to nudge men and boys toward more care work to achieve a more equal distribution of responsibilities? „ What security risks do women face traveling to and from the locations at which they will work or conduct their livelihoods activities? How can these risks be mitigated? Additional questions relevant to public works programs „ Are there specific types of public works programs that are more sensitive to women’s needs and constraints? Can the location of public works programs be chosen in a way that reduces women’s exposure to opportunistic harassment or assault during work or travel to or from the work site? „ What are the opportunities and challenges in the effort to raise the number of women public works supervisors including mitigating risks of their SH? „ What accommodations can be made for women who are pregnant, breastfeeding, or responsible for the care of small children? Is the provision of childcare at the work site feasible? „ Are private, safe, sex-segregated sanitation facilities available at work sites? „ What monitoring or reporting mechanisms are needed to prevent SEA by workplace supervisors or opportunistic harassment or assault by men coworkers? „ Can attendance and task completion verification procedures be designed to reduce the risk of SEA by workplace supervisors? Women active in public works raise many issues, such as the physical demands of the work and sex- based differences in physical capacity and the social norms restricting women’s work versus the opportunity to challenge gender occupational segregation. In some contexts, there is a strict cultural division of the types of labor that are acceptable for men and women to perform, unrelated to physical demands.89 In this case, a balance will have to be struck between nudging the norms toward wider acceptance of women’s engagement in nontraditional work and the need to prevent backlash within households and communities. In some public works programs, there has also been an emerging recognition that some of the basic services provided through the programs may be performed by women (Good Practice Box 16). Engagement with men and traditional leaders during program outreach is an important opportunity to build acceptance for new forms of women’s work. n public works programs, the type of work, the wages and hours, the seasonality of the work, and the potential difficulties in childcare and 89 I supervision arrangements may influence the risk of GBV (Peterman et al. 2017). 62 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE GOOD PRACTICE BOX 16. ADAPTING PUBLIC WORKS TO WOMEN’S NEEDS IN ETHIOPIA In Ethiopia’s Productive Safety Net Program, there has been an effort to ensure that women and men participate in and benefit from the program equally. Since program launch in 2006, several measures have been introduced to address women’s specific needs, as follows: „ Women can opt to perform lighter work than men, and women are encouraged to take on the role of public works team leaders. „ Women who are breastfeeding or who have children ages under 12 months, may participate in training focused largely on nutrition instead of participating in public works. „ A cap on labor that prevented any adult from working more than 20 days per month in public works was reduced to 15 days to encourage women’s participation.. Women thus receive the same monthly wages as men for less work in the programs in recognition of the unequal division of care work. „ Woman-headed households and widows with permanent health and physical challenges receive direct support without a work requirement. „ Public works activities are located no more than 2-kilometers from women’s homes, and childcare centers have been established at work sites. Women face various practical constraints to participation in public works or livelihood activities related to the traditional gender division of household labor. Because women generally have primary responsibility for the majority of care-related activities, they often prefer flexible work hours, task- based work, or piece work for which they are paid based on the completion of discrete tasks rather than daily or hourly wages in public works. While livelihood activities provide more flexibility so women may determine their time use, livelihood and public works programs may still increase women’s time poverty and create the potential for backlash if women are perceived to be neglecting their domestic duties. Consultation with women beneficiaries is important in reaching an understanding of the specific needs of women and deriving possible solutions to reduce any conflict between participation in public works or livelihoods programs and domestic responsibilities. This should include discussion of possible childcare solutions, the inclusion of basic services as part of the menu of public works, and efforts to encourage or incentivize men to take greater responsibility for childcare and other domestic work. Special attention should be paid to avoiding encouraging child labor or making school-age children responsible for the care of siblings or the elderly (Good Practice Box 17). Provide 63 GOOD PRACTICE BOX 17. MOBILE CHILDCARE IN BURKINA FASO Public works programs in Burkina Faso supply a key safety net for vulnerable households through temporary jobs. Yet, they may present a difficult choice for mothers: pass up the opportunity to work or leave young children unattended.a The Youth Employment and Skills Development Project found that many young mothers working at construction sites could not find childcare services; so, they brought their children with them to work and left them near dangerous work areas. This led to an innovative proposal: mobile crèches—mobile daycare groups—that would follow the women as they moved from one work site to another. Offering safe shelter and meaningful stimulation for children of public program workers, the crèches were usually set up under a tree, in an empty building provided by local authorities, or in large weather-resistant tents. The crèches also offered a platform to reach vulnerable children with vaccinations and other health services as well as to share with their parents information on nutrition, hygiene, breastfeeding, and so on. Moreover, they enabled mothers to take advantage of public works opportunities and even created new jobs because childcare providers received the same wages as other workers participating in the public works program. Early lessons indicate that the model can be replicated in different contexts, such as agricultural programs, refugee camps, and outdoor work sites in general. Cameroon and Madagascar have already implemented mobile crèches in selected work sites and plan to scale up program-wide, while programs in the Democratic Republic of Congo and Ethiopia have launched local adaptations of the model recently. a. World Bank (2020a). Participation in public works or economic inclusion programs can expose women to risks of opportunistic GBV during travel to and from work sites or markets or of SEA or harassment by training providers, workplace supervisors, or other men at work sites. Locating work sites close to the homes of women workers, scheduling work to avoid travel during the dark hours, and encouraging and facilitating women’s travel in groups can reduce the risks of travel and may also reduce the risk of opportunistic harassment or assault. Regular work site monitoring, training on codes of conduct, and enforcement of accountability mechanisms (including sanctions) are key strategies for reducing the risk of SEA and harassment of women beneficiaries at work sites. Work sites themselves can also be made safer for women by ensuring that they are not isolated, that safe and, if possible, sex-segregated sanitary facilities are available, that women have access to appropriate work tools, and that work safety measures have been established. Women beneficiaries of public works programs sometimes prefer payment methods other than daily wages, such as task-based wages, that allow them more flexibility to balance work and domestic duties. Task-based payment systems that permit members of a household to share the work of completing a task may be preferable to requirements that women be regularly present at work sites. Piece work is another attractive possibility. For example, payment based on the number of square meters of road cleared may be preferable among women because it would allow them to choose the amount of time they spend on the task on a given day. Microcontracts might also be offered so that women can perform tasks independently. However, care needs to be taken on how the payment for piece work is calculated. Otherwise, women may be exploited into working long hours for low compensation unless work norms and associated payments are precisely specified. Public works programs present an opportunity to nudge norms closer to equal pay for equal work, as well as challenging norms that lead to occupational segregation. 64 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE DESIGN AND IMPLEMENTATION TIPS Tips that are relevant for all economic inclusion programs „ Identify any social constraints on the types of tasks or economic activities women can do and the ways that exist to challenge them. „ Design a communication strategy to address and prevent potential backlash for perceived transgressions of gender or other social norms in women’s work and to nudge these norms away from occupational segregation and toward equal pay for equal work. „ Engage women in the identification of needs and in accommodating their needs to balance work and domestic responsibilities. Additional tips for public works „ Ensure equal pay for equal work, and that any sex segregation of tasks does not lead to women systematically earning less than male counterparts. „ If choosing a piece wage public works payment structure (i.e. payment structure that reimburses beneficiaries for completion of discrete tasks rather than daily wages), ensure that rates reward tasks performed by women and men equitably and proportionally to the effort and time required. „ Consider expanding the menu of work options to include services which are compatible with care work or social norms. „ Consult beneficiaries on how best to adapt work requirements and schedules to accommodate care work. „ Take steps to avoid requiring women having to travel to work sites through insecure areas or outside of daylight hours. „ Identify and facilitate safe childcare options that do not transfer responsibility to older children. Consider establishing mobile or community creches, an option to meet work requirements through provision of childcare for other program participants, or other childcare arrangements. Identify opportunities and incentives to encourage men and boys in sharing of childcare and other care work. „ Accommodate the needs of women who are pregnant or breastfeeding, including suspending work requirements, allowing households to nominate substitutes, or offering the opportunity to substitute work requirements with participation in training activities on issues such as nutrition and early childhood development. „ Ensure access to safe, private, and where possible, sex-segregated sanitation facilities. „ Facilitate the recruitment of women supervisors, including by ensuring SH mitigation measures are in place and by taking steps to minimize risks of opportunistic harassment or assault of women service providers while engaged in program activities. „ Ensure all contractors, supervisors and beneficiaries understand codes of conduct, and sanctions as well as mechanisms for reporting SEA or other abuses of power (see Section 9.1 Codes of Conduct and Section 9.2 Grievance mechanisms). „ Ensure robust monitoring of work sites to prevent SEA by supervisors or harassment by other program participants. Provide 65 9 MANAGE AND MONITOR This section focuses on program management processes in which there are key entry points for assessing, mitigating and monitoring SEA/SH risks, as well as for tracking improvements toward GBV outcomes. Thus, the focus of this section is narrower than the larger process of beneficiary operations management (sometimes referred to as case management). 9.1 CODES OF CONDUCT A code of conduct, sometimes also called code of ethics, outlines the core values and principles of an organization/company and establishes standards of professional conduct and ethical behavior for all staff. The code of conduct articulates the values the organization wishes to foster in leaders and employees and, in doing so, defines both desired behavior and prohibited conduct. As a result, a written code of conduct can become a benchmark against which individual and organizational performance can be measured. The code of conduct offers mandatory guidelines for all actors on ways to exercise good judgment, expectations related to behavior, and provides practical examples of how employee rules should be applied. The code of conduct should be in full alignment with applicable national and local laws and should complement human resources policies that are already in place. The World Bank has not endorsed a template code of conduct, as codes must be tailored to the relevant context. However, sample codes of conduct may be used as reference documents and can be found here. KEY QUESTIONS „ Does the implementing agency have codes of conduct in place that prohibit SEA/SH, including clear definitions of SEA/ SH, actors’ responsibilities, reporting protocols, accountability mechanisms and sanctions? „ Do all contracts with service providers and contractors that will come into contact with beneficiaries, include codes of conduct that prohibit SEA/SH? „ Is training or cascaded information on the SEA/SH provisions of codes of conduct provided to all implementing agency staff, contractors and volunteers? World Bank-financed projects are required to have codes of conduct in place for mitigating SEA/SH risks.90 The ESF requires that Borrowers have clear SEA/SH codes of conduct for SSN program actors in place with sanctions for failure to comply made explicit. These measures should be widely communicated to all program actors, beneficiaries and communities in program implementation areas. 90 See World Bank (2021) and references therein. 66 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Codes of conduct should be integrated into all staff and consultant contracts and explained to all community-level volunteers. Similarly, codes of conduct that identify monitoring and reporting requirements should be integrated into all contracts with service providers (including memorandums of understanding with UN agencies). Codes of conduct should be enforceable and thus accompanied by an accountability and response framework that explains the rights and protections of beneficiaries, and how SEA/SH cases will be handled and referred to appropriate service providers and outline the potential sanctions for misconduct. All program actors should be provided with training on the standards of professional conduct and ethical behavior mandated by the code of conduct as well as associated sanctions. Codes of conduct help can help promote a culture of respect toward beneficiaries and among coworkers. Thus, a code of conduct should be presented as a positive document articulating the organization’s values and desired behavior from its employees, in addition to behavior that is unacceptable and subject to sanctions. Training on codes of conduct, therefore, offers the opportunity to engage all project actors on potentially sensitive topics such as SEA/SH and to clarify that any allegations of SEA/SH will be taken seriously. DESIGN AND IMPLEMENTATION TIPS According to the forthcoming World Bank codes of conduct brief, a code of conduct should include the following: „ Clear language on the organization’s mission, values and principles, linking them with standards of professional conduct „ Clear language describing behaviors that are considered SEA/SH forms of misconduct, and so on. In terms of SEA and SH, this should include (but is not limited to) prohibiting: ƒ Violence, including sexual and other forms of gender-based violence ƒ SEA ƒ SH ƒ Violence against children „ Sanctions that may apply if an employee is in breach of the code should be proportional to the transgression. Potential sanctions include but are not limited to: ƒ Informal or formal warnings ƒ Loss of salary ƒ Suspension of employment (either administrative leave or without payment of salary) ƒ Termination of employment ƒ Referral to police or other authorities as warranted „ How to report suspected violations in a safe and confidential manner and protections of complainants against reprisals „ A code of conduct may also include additional resources for obtaining additional information or guidance Manage and monitor 67 9.2 GRIEVANCE MECHANISMS Grievance mechanisms are an important component of program management. By giving people the capacity to provide feedback to program administrators, a grievance mechanism provides beneficiaries and the general public with a voice in the program’s administration and performance. Grievances can be related to unclear program guidelines; lack of program awareness resulting from insufficient outreach; potential inclusion and exclusion errors; an unsatisfactory package of benefits and services; problems with the payment of benefits or the provision of services; standards of service; mistreatment by front-line workers; fraud and corruption; or the grievance mechanism itself. Increasingly, grievance mechanisms are also important entry points to identify and respond to program-related GBV (in particular SEA by program actors) and can at times also be a venue for reporting broader categories of GBV incidents. As a result, it is also an important instrument for monitoring SEA/SH risks, and more broadly GBV risks, and a pathway for referring beneficiaries to support services for survivors of GBV. KEY QUESTIONS „ Does the grievance mechanism include multiple channels for beneficiaries or other individuals to report SEA if it occurs? Are there safe channels for program actors to report SH by supervisors or other colleagues? Do these channels respect the privacy of complainants and require their informed consent throughout the grievance mechanism process? „ Are standard operating procedures in place for receiving GBV-related complaints, for referring complainants to GBV services and for processing the complaint? Is the grievance mechanism sensitive to social or other constraints women face in reporting cases of GBV? „ What training will grievance mechanism staff and any community focal points or intermediaries (if designated) require to be able to be able to handle GBV complaints appropriately? „ Have GBV support services been identified in program implementation areas? Has their quality and accessibility been assessed? „ Are there any informal community-based mechanisms in place to address cases of GBV? Do these prioritize the well- being of the survivor? Could they be safely integrated into/linked to the program grievance mechanism? „ How will the grievance mechanism handle GBV complaints unrelated to the program should they be reported? „ Does the grievance mechanism provide adequate safeguards against possible risk of backlash from reporting? Grievance mechanisms are essential instruments for mitigating SEA/SH risks. However, the grievance mechanisms of most SSNs are currently ill-equipped to respond to GBV complaints. They may lack confidential or accessible reporting channels, or staff are not trained on protocols for handing GBV complaint, or referral mechanisms are not in place. In many cases, grievance mechanism officers themselves do not believe that cases of GBV fall within their purview or fear retribution for any involvement in such cases. Grievance mechanism officers of the Temporarily Displaced Persons Emergency Relief Program in Pakistan explained in focus group discussions that they would refer such complaints to the local jirga–a traditional mechanism for the public resolution of disputes composed almost always of men (usually religious or 68 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE tribal leaders) whose proceedings often prioritize mediation over the well-being of the complainant against whom backlash for being perceived to bring shame to their family is a significant risk.91 Incidents of GBV, particularly sexual violence, already tend to be grossly underreported for various reasons.92 These include victims’ fear of backlash, stigmatization or revictimization through forced mediation with the perpetrator; the dearth of quality services to respond to their needs, particularly in low-income countries; and the high rates of impunity of perpetrators. Given such challenges, survivors and witnesses are not likely to come forward and report incidents using a typical SSN program grievance mechanism. To adequately handle reports of GBV, additional considerations are thus necessary in the design or adaptation of SSN program grievance mechanisms. These include identifying appropriate entry points for communicating about the mechanism, identifying trusted reporting channels, and establishing protocols for data sharing, collection, and storage. Mechanisms are needed that create safe, enabling spaces for survivors to report GBV incidents and offer a safe, ethical, survivor-centered response in which the safety and well-being of the SEA/SH survivor is the first priority and any action is only taken with the survivor’s informed consent. The grievance mechanism should offer survivors “warm referrals” explaining why the service can be helpful for their specific needs and actively helping them access the referral (e.g., by offering to make a call on their behalf or to accompany them to the service provider). Cross-sectoral coordination within the World Bank portfolio as well as with development partners in each country is also important in determining the role and scope of the SSN program’s grievance mechanism with regard to GBV. In some cases, there may be grievance mechanisms linked to other national programs or World Bank projects better equipped to handle GBV complaints to which the SSN grievance mechanism could be linked. It is important to have a map of GBV services (and their quality) to which complainants can be referred. This mapping is best done at the country or portfolio level and then made available to all project teams. In many cases, this information is already being collected by other organizations and agencies including government agencies responsible for the provision of social services and by the in-country GBV working groups chaired by the United Nations Population Fund.93 A mapping exercise of GBV services led by the South Asia gender team is being used to build an interactive online tool for project teams and clients, and is a best practice example.94 In FCV settings, humanitarian implementers will likely have mapped services (or their absence) as part of their assessment of protection needs. If preexisting information is not available, it is recommended that teams conduct a project-specific mapping exercise in program implementation areas when residual risks of SEA/SH are considered substantial or high. All mapping exercises should take into account the needs of minority groups, accessibility needs for persons with disability, and any other specific constraints to access. The ESF requires that projects have a “grievance mechanism that will be proportionate to the risks and impacts of the project” and that, at a minimum, the grievance mechanism be able to respond to allegations of SEA/SH by program actors. Whether or not to manage allegations of GBV against beneficiaries indirectly related to the program (such as IPV or abuses committed by other service providers) will depend on the context and risks related to each program. A wider scope might be relevant in fragile or conflict-affected contexts, for programs reaching particularly vulnerable populations, or 91 Durrani et al. (2020). 92 Palermo, Bleck, and Peterman (2014) estimate that only 7 percent of GBV experiences are formally reported, according to Demographic and Health Surveys data from 24 low- and middle-income countries. 93 The United Nations Population Fund normally chairs a GBV working group involving key partners and civil society organizations responsible for service provision. A mapping of programs is available through this coordination group and regularly updated. 94 The mapping focused on services that respond to GBV in Afghanistan, Pakistan, Sri Lanka, Maldives, Bangladesh, Bhutan, Nepal and selected priority states in India. Click here for more information. Manage and monitor 69 when the program includes accompanying measures focusing on GBV prevention that may lead to backlash if not properly managed (Boxes 18 and 19). In these circumstances, additional mechanisms should be put in place to handle allegations of GBV even when not directly involving program actors. In these instances, grievance mechanism staff will need to be trained on protocols to handle SEA/SH complaints but also to receive and refer other GBV complaints. See Annex 3 for SEA/SH risk mitigation measures for World Bank Human Development investment projects. GOOD PRACTICE BOX 18. STRENGTHENING SOUTH SUDAN’S GRIEVANCE MECHANISM FOR GBV MITIGATION The Safety Net and Skills Development Project in South Sudan provides useful lessons on how the grievance mechanism can be strengthened for GBV risk mitigation. A 2019 process evaluation revealed a series of barriers to women’s access to the initial community-based grievance mechanism, particularly in terms of reporting grievances relating to GBV. The fact that Appeals Committees were selected based on male-dominated traditional power structures made women reticent to approach the Group Leader or Appeals Committee in charge of processing grievances. Reporting was further impeded by the widespread belief that family or personal problems such as domestic violence should be addressed within the family and that, consequently, submitting GBV-related grievances would be seen as violating this norm and reflect negatively on the complainant. The assessment also discovered that there had been limited sensitization efforts regarding program-related GBV risks in 3 of the 4 research locations and that the program’s grievance mechanism did not include GBV/SEA reporting mechanisms or referral pathways for beneficiaries. As a result, gender-responsive mechanisms were put in place to respond to experience of GBV associated with the program, such as introducing other traditional and community-based channels for handling GBV incidents and sharing information with beneficiaries about relevant services offered by partners. The grievance mechanism was thus identified as a good entry point to raise awareness more broadly on GBV/SEA issues and promote women’s rights: empowering women and giving them information to seek support in a confidential manner are important first steps in changing community norms and behaviors. The choice of the appropriate model and scope of the grievance mechanism will depend both on the assessment of risks of program-related GBV and on the capacity of the broader grievance mechanism. SSN programs can consider different arrangements to handle program-related GBV allegations: (i) adapting the program grievance mechanism to allow for the uptake of GBV allegations, (ii) linking the program grievance mechanism with an existing intermediary to handle such allegations; or (iii) building an independent SEA/SH grievance mechanism and outsourcing the role of a third party or specialized team within the program management.95 While grievance mechanisms establish formal channels through which SSNs respond to reports of GBV, particularly cases of SEA/SH, it is also important to raise the general awareness of all program actors who interact with beneficiaries. Program actors should be provided with guidance on what to do in case they become aware of any incidents of program-related GBV. Program actors should be advised not to take action reporting incidents of GBV on behalf of beneficiaries without their consent, and should instead be advised to direct beneficiaries who so wish to seek help through the grievance mechanism. Without disclosing any identifying information regarding specific incidents of GBV without informed consent of survivors, which is not permissible, program actors should learn how signal emerging risks or localized problems to their management, so that appropriate program-level responses can be implemented. 95 For more detailed guidance, see World Bank (2020b). 70 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE GOOD PRACTICE BOX 19. ESTABLISHING A GBV-SENSITIVE GRIEVANCE MECHANISM IN ZAMBIA The GEWEL Project in Zambia has pioneered the use of the grievance mechanism to receive and refer GBV allegations. GEWEL aims to enhance girls’ and women’s well-being through secondary education and livelihoods support, but as with any intervention it is possible that their participation puts them at increased risk of GBV (e.g., adolescent girls living in informal or semi-formal boarding facilities). To ensure safety and mitigate these risks, GEWEL has supported the Government to put in place a comprehensive GBV Action Plan, a key element of which is the accelerated roll-out of a GBV-sensitive grievance mechanism. Key elements of Zambia’s GBV-responsive grievance mechanism: „ Multiple reporting channels: In addition to the complaint boxes already included in the initial grievance mechanism, two additional channels were established to facilitate the submission of GBV allegations. First, women focal points were selected in each community (based on a set of criteria, including being a trusted and respected member of the community) to lead sensitization efforts and collect complaints, especially when the complainer wishes to remain anonymous or when they have insufficient literacy skills to submit a written complaint by themselves. Second, GEWEL hired a specialized NGO, Lifeline/Childline Zambia, to provide access to its existing national hotline for GBV and other child protection issues. Callers receive counselling over the phone, referral to appropriate services in their district, and case management until the case is closed. „ Survivor safety and robust monitoring and evaluation (M&E): Important steps were taken in the design of the grievance mechanism monitoring and information system module to ensure survivor safety including limited access to the module and serious complaint information not being visible to all users. To reinforce data privacy and security, a simple information sharing protocol was developed, outlining guiding principles for data sharing, management and security. The monitoring and information system module also includes an interactive M&E framework where officers at various levels can design, carry out, or review M&E activities and reports in real time. „ Financial support to survivors: The GEWEL grievance mechanism is setting up a Fund to Address Serious Complaints that is allocated to each district. Guidelines are currently being developed to ensure an immediate response to GBV cases at the district level as well as providing additional support needed by survivors. Lessons learned While still in the early stages of roll-out, several lessons have been learned which could help those seeking to design or adapt their grievance mechanism for GBV, including: „ Training for grievance mechanism officers should include sessions on GBV, including SEA/SH. For some officers, it will be their first time being exposed to these concepts and therefore significant time should be allocated for these sessions. „ It is important to work toward a human development practice-wide approach from the start. Ideally, health, education, and social protection work together on a harmonized approach to GBV grievances. Services for GBV response are limited in availability and quality on the ground in many developing countries. While an SSN program may not be directly responsible for or able to put these services in place, there is a need to advocate for the strengthening of national systems. „ GBV is a long-term development issue in the human development sector, not just a risk to be managed within each program. As such, it takes time to establish functional GBV-responsive grievance mechanisms for programs operating at a national level. „ It is possible to adapt grievance mechanisms to shocks such as COVID-19 by making modifications to the placement of boxes and complaints forms, strengthening radio messaging and other innovative forms of communication, and coordination with women’s organizations at community level. Manage and monitor 71 DESIGN AND IMPLEMENTATION TIPS96 „ Adopt a survivor-centered approach whereby the safety and well-being of the survivor is the first priority and any action is only be taken with the survivor’s informed consent. „ Train grievance mechanism staff on protocols for handling GBV-related reports, emphasizing survivor-centric approaches and the importance of maintaining confidentiality. „ Involve communities, local authorities and women’s groups in the development of grievance mechanism policies and procedures. „ Undertake community outreach to explain the functioning of the grievance mechanism. Disseminate information on how to report complaints, including that reporting grievances will not negatively affect program participation. „ Include multiple channels through which complaints can be registered in a safe and confidential manner, including through anonymous complaints reporting mechanisms or trusted intermediaries (ideally, trusted women). „ Communicate any mandatory reporting requirements before the filing of a formal grievance. Information on relevant legislation mandating that the grievance mechanism reports SEA/SH (or other types of GBV) allegations should be delivered to survivors early on and prior to any disclosure to respect the survivor-centered approach. „ Avoid storing any identifiable information on the survivor in the grievance mechanism and keep all information confidential. The grievance mechanism should not require disclosure of, or record, information on aspects of the GBV incident other than (a) the nature of the complaint, (b) if, to the best of their knowledge, the perpetrator was associated with the program, and (c) if possible, the age and sex of the survivor. „ Design the grievance mechanism to allow for the immediate referral of survivors to GBV service providers, by prior mapping of services and the establishment of fast-tracked procedures. Ensure all program actors are trained on these procedures. Consider providing monetary or other assistance to facilitate survivors’ access to GBV support services. „ Consider developing an information sharing protocol that is endorsed by all relevant stakeholders. This may cover (a) guiding principles for data sharing, data management and security; (b) internal information sharing procedures; (c) accountability; (d) timelines; (e) breaches. 9.3 MONITORING AND EVALUATION M&E systems play a central role in tracking the performance of SSN activities and are important channels for providing feedback for continuous improvements. Regular data collection is an important element of program management and a useful tool for monitoring GBV risk levels as well as the performance of mitigation measures such as the grievance mechanism. However, given that specific expertise and safeguards must be in place to collect data on GBV, it is generally not advisable to measure GBV rates as part of regular program monitoring systems. Evaluation is a key element of program oversight that allows policy makers and mangers to assess the effectiveness and efficiency of a program in relation to both processes and impacts. Evaluations provide important information for making course corrections including those that minimize any program-related GBV risks or that enhance positive impacts. 96 See World Bank (2020b) for more detailed guidance. 72 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE KEY QUESTIONS „ If reducing GBV more broadly is a program objective, how will progress toward this objective be measured given the sensitivity around GBV-related data collection? „ Are there opportunities to gather beneficiary feedback related to the impact pathways outlined in Section 2 that could help track improvements in women’s empowerment and reduction in risks of GBV? „ Are the program activities or adaptations aimed at mitigating GBV risk or responding to GBV being implemented as intended? „ What output and outcome indicators can be included in the results framework to monitor implementation of any program activities or adaptations aimed at preventing GBV? „ Is there scope to complement routine M&E data collection with assessments to gauge the quality of implementation (e.g. spot checks, focus group discussions with beneficiaries)? „ If planning an impact evaluation, are there particular risks or pathways elements which should be measured (see Box 3)? The main role of the M&E is to track whether program activities, including any activities focused on mitigating or preventing GBV, are implemented as intended and to identify areas for improvement. Continuous and robust monitoring is critical for ensuring that design choices are effective, and that program actors correctly understand and carry out their responsibilities. For instance, establishing a gender-sensitive grievance mechanism may not achieve its desired impact if not accompanied by a robust sensitization campaign to inform beneficiaries about the various complaints channels or if grievance mechanism officers do not handle complaints following established protocol. Indicators should, therefore, include the number of complaints received (disaggregated by type of complaint and gender of plaintiff), share of complaints responded to, time taken to complete each step of the complaint processing process, and so on. Including specific output and outcome indicators related to GBV mitigation and prevention in the M&E framework not only provides a tool for tracking implementation but may also strengthen implementation by flagging its importance to program staff. In addition, effective tracking of program-related GBV complaints and grievance mechanism responses is essential for effective risk mitigation. The implementing agency is accountable for SEA/SH and broader program-related GBV risk management; for establishing mitigation measures (as reflected in the environmental and social management framework and the environmental and social management plan); and for regularly monitoring of the occurrence of SEA/SH or other GBV-related cases as well as the implementation of the mitigation measures identified during the SEA/SH and GBV risk screening. It is recommended that teams select indicators of key mitigation measures for inclusion in the project’s Results Framework (see Box 18 for guidance on types of monitoring indicators). According to the principle of proportionality, the type and timeframe of the system in place to monitor response to SEA/SH complaints will be determined by the level of residual project-specific SEA/SH risk. Given heightened risks in FCV environments, it is particularly important to establish robust GBV risk monitoring mechanisms in coordination with any other actors working in the field. Monitoring activities should not include the collection on experiences of GBV directly from beneficiaries, as this should only be done following international guidelines on the safe and ethical collection of data on GBV. Rather, data to measure program impacts on GBV prevalence could be collected in the context of an independent impact Manage and monitor 73 evaluation that adheres to ethical protocols around collecting sensitive information and is able to refer respondents to appropriate survivor support services. While project-related GBV risks and mitigation measures should be identified during the design stage, new (or missed) risks may emerge during implementation. Periodic revision of monitoring tools should be part of regular supervision to reflect risks that may emerge during implementation, followed by the establishment of additional mitigation measures as relevant. New GBV risks may emerge at any point during implementation and thus flexibility to address unforeseen consequences should be built into the design of M&E framework. This is particularly relevant in high-risk contexts, where teams should regularly monitor existing sources on GBV rates, including official GBV rates reported by government agencies, social services, police and the judiciary.97 Regular exchanges of information should also be established with other organizations and agencies operating in program implementation areas and conducting periodic evaluations of GBV risks. Innovations in monitoring metadata from various social media platforms (e.g. statistics re. use of hashtags, Facebook post, and so on) can be another source of information particularly on attitudes toward GBV.  owever, GBV is significantly underreported everywhere. The rates of female homicides is one of the more reliable indicators, but represents 97 H only the most extreme form of violence against women. 74 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE BOX 20. TYPES OF MONITORING INDICATORS M&E plans and frameworks should include ways to assess whether GBV mitigation measures are implemented as intended and to continuously observe and investigate potentially adverse reactions to the program. Four main types of M&E indicators should be considered: 1. Output-level indicators to determine whether program activities, including those aimed at mitigating or reducing risk of GBV, are being implemented. For example, output indicators include number of women program staff and front-line service providers recruited and retention rates, number of program staff who have signed codes of conduct and received training on SEA/SH, number of community meetings conducted as part of inclusive outreach efforts on SEA/SH, or number of grievance mechanism focal points trained, percentage of public work sites with access to private, safe and sex-segregated sanitation facilities, numbers of women work site supervisors, and so on. 2. Outcome-level indicators to assess whether program activities achieved the benefits they were designed to deliver. Examples of outcome indicators include share of women participants, the number of GBV-related complaints handled and resolved through the grievance mechanism, the changes in perceptions or knowledge from GBV or gender-specific accompanying measures, the share of women among program recipients, increases in rates of access to financial services, mobile phones, digital banking, beneficiary perceptions of security while traveling to/from or during participation in program activities, percentage of beneficiaries of economic inclusion programs with access childcare, availability of sex-disaggregated public works program, sex-disaggregated monitoring of public works wage rates and total earnings, and so on. 3. Quality of implementation indicators to go beyond measuring outputs and assessing whether the way in which program activities are carried out meets certain standards: obtaining the more granular, potentially intangible, information on quality of implementation relies on qualitative data collection. Under the M&E plan, this typically includes observations during regular field visits, semi-structured interviews with front-line service providers and beneficiaries during spot checks, or larger beneficiary engagement surveys. While no direct questions on experience of GBV should be included, as discussed earlier, these instruments can be critical to understand the quality of implementation, including measures around the quality of information on the grievance mechanism, the availability of the grievance mechanism officers and their professionalism, the quality of the content of accompanying measures, the regularity of payments, the respect of good practice around scheduling of payments, the respect of the community involvement in the intake/registration period, and so on. For example, this can include questions to men and women beneficiaries about changes in household decision making, how conflicts are resolved, how decisions about spending SSN resources are made, or self- reported stress and well-being. They can also include questions to women about connections to other women in the community, and feelings of agency. 4. Impact indicators try and measure the changes observed as a result of the program implementation. Impact evaluations might want to measure GBV rate following international standards of ethical research.a Other measures of impact can also be assessed through quantitative or qualitative assessments, related to the impact pathways described in Section 2. This could include, for instance, whether women feel more empowered in various dimensions, whether the program is building women’s social capital, whether social norms have evolved and perceptions changed, whether knowledge has evolved, whether financial inclusion has been deepened, and so on. Community engagement is also useful for continuously gathering information about how the program is affecting the community and potential backlash against women beneficiaries. Regular consultations with beneficiaries and local women’s organizations should be conducted to identify GBV risks and adaptations to program activities and processes to reduce these risks. Any public dialogues on GBV risk or other potentially sensitive topics should be held to ensure women’s safety and privacy. a. Ellsberg and Heise (2005). Manage and monitor 75 Depending on the scope and capacity of each program, routine M&E data collection may be accompanied by periodic assessments to measure the quality of implementation. This may include spot checks and field visits by program staff to observe implementation and to get feedback from program beneficiaries and front-line service providers directly. As well, qualitative research, such as focus group discussions with beneficiaries, conducted independently to provide insight into gender dynamics and GBV drivers in areas of program implementation or gain insight into the effects of the program or specific program activities. Process evaluations may also be useful for more systematically assessing whether activities designed to reduce GBV risk are being implemented as planned. Ideally, they should be conducted relatively early during program implementation to allow for timely identification of bottlenecks and course correction. Increased attention to GBV and women’s participation in the program requires M&E staff engaging in supervision and program implementation with gender expertise and capacity to monitoring GBV- related indicators. Capacity building of program actors with M&E responsibilities should include training on gender-sensitive implementation, including on handling potentially sensitive data collection related to GBV and the importance of respecting beneficiaries’ privacy and confidentiality. To the extent possible, programs should prioritize recruiting M&E specialists and officers with gender expertise. As well, efforts should be made to recruit and retain women staff at all levels of program implementation. This requires understanding any constraints they may face and taking measure to reduce these. It is also important to establish a safe and respectful work environment in which any SH complaints are taken seriously. DESIGN AND IMPLEMENTATION TIPS „ Include output and outcome indicators in the results framework to monitor the implementation of GBV-related program activities and of all mitigation measures (see Box 18). „ Leverage project cycle decision points, such as the mid-term review, to reassess GBV risks and to make adjustments to program design and mitigation measures as needed. „ Do not measure prevalence of GBV as part of regular monitoring activities as there are significant risks to women unless data on GBV is collected by trained experts following established safety and ethics guidelines.98 „ Consider building into the project’s M&E regular community feedback mechanisms that are accessible to women to allow for the safe disclosure of any program-related GBV concerns. „ Build the capacity of M&E staff on handling GBV-related data and sensitize them to the importance of implementing and monitoring activities to mitigate or reduce GBV. „ Consider conducting GBV-focused impact evaluations following all ethical protocols to measure the program’s effect on GBV if the context or beneficiary profile indicates a high risk. „ Monitor incidents SEA/SH and other forms of program-related GBV incidents reported through grievance mechanisms. 98 Ellsberg and Heise (2005). 76 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE BIBLIOGRAPHY Abramsky, Tanya, Karen Devries, Ligia Kiss, Janet Nakuti, Nambusi Kyegombe, Elizabeth Starmann, Bonnie Cundill, et al. 2014. “Findings from the SASA! Study: A Cluster Randomized Controlled Trial to Assess the Impact of a Community Mobilization Intervention to Prevent Violence against Women and Reduce HIV Risk in Kampala, Uganda.” BMC Medicine 12 (July), article 122. https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-014-0122-5. AFI (Alliance for Financial Inclusion). 2019. KYC Innovations, Financial Inclusion, and Integrity In Selected AFI Member Countries. AFI Guideline Note 32 (March). Kuala Lumpur, Malaysia: AFI. Agarwal, Bina, and Pradeep Panda. 2007. “Toward Freedom from Domestic Violence: The Neglected Obvious.” Journal of Human Development and Capabilities 8 (3). 359–88. Ahmed, Akhter U., Agnes R. Quisumbing, Mahbuba Nasreen, John F. Hoddinott, and Elizabeth Bryan. 2009. Comparing Food and Cash Transfers to the Ultra Poor in Bangladesh. Research Monograph 163. Washington, DC: International Food Policy Research Institute. Aker, Jenny C., Rachid Boumnijel, Amanda McClelland, and Niall Tierney. 2016. “Payment Mechanisms and Antipoverty Programs: Evidence from a Mobile Money Cash Transfer Experiment in Niger.” Economic Development and Cultural Change 65 (1): 1–37. Akresh, Richard, Damien de Walque, and Harounan Kazianga. 2013. “Cash Transfers and Child Schooling: Evidence from a Randomized Evaluation of the Role of Conditionality.” Policy Research Working Paper 6340, World Bank, Washington, DC. Ali, Snigdha and Stephanie Kuttner. 2020. “Safety First: Social Safety Nets and Preventing Gender-Based Violence; Case Study of Bangladesh ISPP and EGPP Programs.” World Bank, Washington, DC. Almås, Ingvild, Alex Armand, Orazio P. Attanasio, and Pedro Carneiro. 2018. “Measuring and Changing Control: Women’s Empowerment and Targeted Transfers.” Economic Journal 128 (612): F609–F639. Amaral, Sofia, Siddhartha Bandyopadhyay, and Rudra Sensarma. 2015. “Employment Programmes for the Poor and Female Empowerment: The Effect of NREGS on Gender-Based Violence in India.” Journal of Interdisciplinary Economics 27 (2): 199–218. Angelucci, Manuela. 2008. “Love on the Rocks: Domestic Violence and Alcohol Abuse in Rural Mexico.” B.E. Journal of Economic Analysis and Policy 8 (1): 1–43. Armand, Alex, Paul Atwell, and Joseph F. Gomes. 2020. “The Reach of Radio: Ending Civil Conflict through Rebel Demobilization.” American Economic Review 110 (5): 1395–1429. Ashraf, Nava, Dean S. Karlan, and Wesley Yin. 2010. “Female Empowerment: Impact of a Commitment Savings Product in the Philippines.” World Development 38 (3): 333–44. Bibliography 77 Bachas, Pierre, Paul Gertler, Sean Higgins, and Enrique Seira. 2018. “Digital Financial Services Go a Long Way: Transaction Costs and Financial Inclusion.” AEA Papers and Proceedings 108 (May): 444–48. Banerjee, Abhijit Vinayak, Esther Duflo, Nathanael Goldberg, Dean S. Karlan, Robert Osei, William Parienté, Jeremy Shapiro, Bram Thuysbaert, and Christopher Udry. 2015. “A Multifaceted Program Causes Lasting Progress for the Very Poor: Evidence from Six Countries.” Science 348 (6236): 772–89. Banerjee, Abhijit Vinayak, Eliana La Ferrara, and Victor Orozco. 2019. “Entertainment, Education, and Attitudes toward Domestic Violence.” AEA Papers and Proceedings 109 (May): 133–37. Baranov, Victoria, Lisa Cameron, Diana Contreras Suarez, and Claire Thibout. 2021. “Theoretical Underpinnings and Meta-analysis of the Effects of Cash Transfers on Intimate Partner Violence in Low- and Middle-Income Countries.” Journal of Development Studies 57 (1): 1–25 Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, and Tanja Schmidt. 2016. Cash Transfers: What Does the Evidence Say? A Rigorous Review of Programme Impact and of the Role of Design and Implementation Features. With Luca Pellerano. July. London: Overseas Development Institute. Bastian, Gautam, Markus P. Goldstein, and Sreelakshmi Papineni. 2017. “Are Cash Transfers Better Chunky or Smooth? Evidence from an Impact Evaluation of a Cash Transfer Program in Northern Nigeria.” Gender Innovation Lab Policy Brief 21, World Bank, Washington, DC. Benhassine, Najy, Florencia Devoto, Esther Duflo, Pascaline Dupas, and Victor Pouliquen. 2015. “Turning a Shove into a Nudge? A ‘Labeled Cash Transfer’ for Education.” American Economic Journal: Economic Policy 7 (3): 86–125. Bilali, Rezarta, and Johanna Ray Vollhardt, and Jason Rarick. 2016. “Assessing the Impact of a Media- Based Intervention to Prevent Intergroup Violence and Promote Positive Intergroup Relations in Burundi.” Journal of Community and Applied Social Psychology 26 (3): 221–35. Bloch, Francis, and Vijayendra Rao. 2002. “Terror as a Bargaining Instrument: A Case Study of Dowry Violence in Rural India.” American Economic Review 92 (4): 1029–43. Bobonis, Gustavo J., Melissa González-Brenes, and Roberto Castro. 2013. “Public Transfers and Domestic Violence: The Roles of Private Information and Spousal Control.” American Economic Journal: Economic Policy 5 (1): 179–205. Bohnet, Iris. 2016. What Works: Gender Equality by Design. Cambridge, MA: Belknap Press of Harvard University. Brody, Carinne., Thomas de Hoop, Martina Vojtkova, Ruby Warnock, Megan Dunbar, Padmini Murthy, and Shari L. Dworkin. 2015. “Economic Self-Help Group Programs for Improving Women’s Empowerment: A Systematic Review.” Campbell Systematic Reviews 11 (1): 1–182. Buller, Ana Maria, Melissa Hidrobo, Amber Peterman, and Lori L. Heise. 2016. “The Way to a Man’s Heart Is through His Stomach? A Mixed Methods Study on the Causal Mechanisms through Which Cash and In-kind Food Transfers Decreased Intimate Partner Violence.” BMC Public Health 16 (488): 1–13. 78 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Buller, Ana Maria, Amber Peterman, Meghna Ranganathan, Alexandra Bleile, Melissa Hidrobo, and Lori L. Heise. 2018. “A Mixed-Method Review of Cash Transfers and Intimate Partner Violence in Low- and Middle-Income Countries.” World Bank Research Observer 33 (2): 218–58. https://doi.org/10.1093/wbro/lky002. CaLP (Cash Learning Partnership). 2018. Collected Papers on Gender and Cash Transfer Programmes in Humanitarian Contexts. Oxford, UK: CaLP. Camacho, Adriana, Alejandro Gaviria, and Catherine Rodríguez. 2016. “Drug Consumption in Colombia.” Documento CEDE 2016-36 (November 28), Center for Economic Development Studies, Department of Economics, Universidad de Los Andes, Bogotá, Colombia. Cheung, Maria. 2012. “Edutainment Radio, Women’s Status, and Primary School Participation: Evidence from Cambodia.” Research Paper in Economics 5 (April 10), Department of Economics, Stockholm University, Stockholm. Corboz, Wahid Siddiq, Osman Hemat, Esnat D. Chirwa, and Rachel Jewkes. 2019. “What Works to Prevent Violence against Children in Afghanistan? Findings of an Interrupted Time Series Evaluation of a School-Based Peace Education and Community Social Norms Change Intervention in Afghanistan.” PLOS ONE 14 (8): e0220614. Cross, Allyson, Tenzin Manell, and Melanie Megevand. 2018. “Humanitarian Cash Transfer Programming and Gender-Based Violence Outcomes: Evidence and Future Research Priorities.” Report (November), Women’s Refugee Commission and International Rescue Committee, New York. Cullen, Claire. 2020. “Method Matters: Underreporting of Intimate Partner Violence in Nigeria and Rwanda.” Policy Research Working Paper 9274, World Bank, Washington, DC. Cullen, Claire, Paula Lorena Gonzales Martinez, and Sreelakshmi Papineni. 2020. “Empowering Women without Backlash? Experimental Evidence on the Impacts of a Cash Transfer and Community Livelihoods Program on Intimate Partner Violence in Northern Nigeria.” Working paper, Africa Gender Innovation Lab, World Bank, Washington, DC. Dervisevic, Ervin, Elizaveta Perova, and Abhilasha Sahay. 2021. “Long-Term Impacts of a Short Exposure to CCTs in Adolescence: Evidence from the Philippines.” Working paper, World Bank, Washington, DC. Dooley, Meagan, Abby Fried, Ruti Levtov, Kate Doyle, Jeni Klugman, and Gary Barker. 2019. Applying a Masculinities Lens to the Impacts of Social Safety Nets on Women’s Economic Empowerment. EBA Report 2019:07. Stockholm: Expert Group for Aid Studies. Doyle, Kate, Ruti G. Levtov, Gary Barker, Gautam G. Bastian, Jeffrey B. Bingenheimer, Shamsi Kazimbaya, Anicet Nzabonimpa, et al. 2018. “Gender-Transformative Bandebereho Couples’ Intervention to Promote Male Engagement in Reproductive and Maternal Health and Violence Prevention in Rwanda: Findings from a Randomized Controlled Trial.” PLOS ONE 13 (4): e0192756. Dupas, Pascaline, and Jonathan Robinson. 2013. “Savings Constraints and Microenterprise Development: Evidence from a Field Experiment in Kenya.” American Economic Journal: Applied Economics 5 (1): 163–92. Bibliography 79 Durrani, Ahmad Shah, Stephanie Kuttner, and Mina Zamand. 2020. “Safety First: Social Safety Nets and Gender-Based Violence; Case Study of the FATA-TDP Emergency Recovery Project, Pakistan.” World Bank, Washington, DC. Ellsberg, Mary Carroll, and Lori L. Heise. 2005. Researching Violence against Women: A Practical Guide for Researchers and Activists. Washington, DC: Program for Appropriate Technology in Health; Geneva: World Health Organization. https://www.who.int/reproductivehealth/publications/ violence/9241546476/en/. Ellsberg, Mary Carroll, Rodolfo Peña, Aandres Herrera, Jerker Liljestrand, and Anna Winkvist. 1999. “Wife Abuse among Women of Childbearing Age in Nicaragua.” American Journal of Public Health 89 (2): 241–44. Ellsberg, Mary, Arango, Diana J., Morton, Matthew, Gennari, Floriza, Kiplesund, Sveinung, Contreras, Manuel, & Watts, Charlotte. 2015. “Prevention of violence against women and girls: what does the evidence say?” The Lancet, 385(9977), 1555–1566. Eswaran, Mukesh, and Nisha Malhotra. 2011. “Domestic Violence and Women’s Autonomy in Developing Countries: Theory and Evidence.” Canadian Journal of Economics 44 (4): 1222–63. Farmer, Amy, and Jill Tiefenthaler. 1997. “An Economic Analysis of Domestic Violence.” Review of Social Economy 55 (3): 337–58. Field, Erica M., Rohini Pande, Natalia Rigol, Simone G. Schaner, and Charity Troyer Moore. 2019. “On Her Own Account: How Strengthening Women’s Financial Control Affects Labor Supply and Gender Norms.” NBER Working Paper 26294 (September), National Bureau of Economic Research, Cambridge, MA. Fulu, Emma, Rachel Jewkes, Tim Roselli, Claudia García-Moreno, and UN Multi-country Cross-Sectional Study on Men and Violence Research Team. 2013. “Prevalence of and Factors Associated with Male Perpetration of Intimate Partner Violence: Findings from the UN Multi-country Cross-Sectional Study on Men and Violence in Asia and the Pacific.” Lancet Global Health 1 (4): e187–e207. Fulu, Emma, Stephanie Miedema, Tim Roselli, Sarah McCook, Ko Ling Chan, Regine Haardörfer, Rachel Jewkes, and UN Multi-country Study on Men and Violence Study Team. 2017. “Pathways between Childhood Trauma, Intimate Partner Violence, and Harsh Parenting: Findings from the UN Multi- country Study on Men and Violence in Asia and the Pacific.” Lancet Global Health 5 (5): e512– e522. García-Moreno, Claudia, Lori L. Heise, Henrica A. F. M. Jansen, Mary Carroll Ellsberg, and Charlotte Watts. 2005. “Violence against Women.” Science 310 (5752): 1282–83. doi: 10.1126/science.1121400. Garz, Seth, Rachel Heath, Elijah Kipchumba, and Munshi Sulaiman. 2020. “Evidence of Digital Financial Services Impacting Women’s Economic Empowerment: What Explains the Impact and What Is Left to Learn?“ September 29, BRAC Institute of Governance and Development, Dhaka, Bangladesh. Gibbs, Andrew, Nata Duvvury, and Stacey Scriver. 2017. “What Works Evidence Review: The Relationship between Poverty and Intimate Partner Violence.” What Works to Prevent Violence against Women and Girls, September, U.K. Department for International Development, London. 80 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Gibbs, Andrew, Jessica Jacobson, and Alice Kerr-Wilson. 2017. “A Global Comprehensive Review of Economic Interventions to Prevent Intimate Partner Violence and HIV Risk Behaviours.” Global Health Action 10 (2): 1290427. Gibbs, Andrew, Laura Washington, Samantha Willan, Nolwazi Ntini, Thobani Khumalo, Nompumelelo Mbatha, Yandisa Sikweyiya, et al. 2017. “The Stepping Stones and Creating Futures: Intervention to Prevent Intimate Partner Violence and HIV-Risk Behaviours in Durban, South Africa; Study Protocol for a Cluster Randomized Control Trial and Baseline Characteristics.” BMC Public Health 17 (1): 336. Glass, Nancy, Nancy A Perrin, Anjalee Kohli, Jacquelyn Campbell, and Mitima Mpanano Remy. 2017. “Randomised Controlled Trial of a Livestock Productive Asset Transfer Programme to Improve Economic and Health Outcomes and Reduce Intimate Partner Violence in a Postconflict Setting.” BMJ Global Health 2 (1): e000165. Green, Donald P., Anna M. Wilke, and Jasper Cooper. 2020. “Countering Violence against Women by Encouraging Disclosure: A Mass Media Experiment in Rural Uganda.” Comparative Political Studies 53 (14): 2283–2320. Green, Eric P., Christopher Blattman, Julian Jamison, and Jeannie Annan. 2015. “Women’s Entrepreneurship and Intimate Partner Violence: A Cluster Randomized Trial of Microenterprise Assistance and Partner Participation in Post-conflict Uganda.” Social Science and Medicine 133 (May): 177–88. Guilbert, Nathalie, and Rachael Pierotti. 2016. “Intra-Household Dynamics and the Design of Social Protection Programs: The Case of Polygamous Households in North Burkina Faso.” Policy Brief 16 (March), Gender Innovation Lab, World Bank, Washington, DC. https://openknowledge. worldbank.org/handle/10986/25448. Haushofer, Johannes, and Ernst Fehr. 2014. “On the Psychology of Poverty.” Science 344 (6186): 862–67. doi: 10.1126/science.1232491. Haushofer, Johannes, Robert Mudida, and Jeremy Shapiro. 2019. “The Comparative Impact of Cash Transfers and Psychotherapy on Psychological and Economic Well-Being.” Working paper, Busara Center for Behavioral Economics, Nairobi, Kenya. Haushofer, Johannes, and Jeremy Shapiro. 2016. “The Short-Term Impact of Unconditional Cash Transfers to the Poor: Experimental Evidence from Kenya.” Quarterly Journal of Economics 131 (4): 1973– 2042. Haushofer, Johannes, and Jeremy Shapiro. 2018. “The Long-Term Impacts of Unconditional Cash Transfers: Experimental Evidence from Kenya.” Working paper, Busara Center for Behavioral Economics, Nairobi, Kenya. Hautzinger, Sarah. 2003. “Researching Men’s Violence: Personal Reflections on Ethnographic Data.” Men and Masculinities 6 (1): 93–106. Heath, Rachel, Melissa Hidrobo, and Shalini Roy. 2020. “Cash Transfers, Polygamy, and Intimate Partner Violence: Experimental Evidence from Mali.” Journal of Development Economics 143 (March): 102410. https://doi.org/10.1016/j.jdeveco.2019.102410. Bibliography 81 Heinrich, Carolyn J., and Matthew T. Knowles. 2020. “A Fine Predicament: Conditioning, Compliance, and Consequences in a Labeled Cash Transfer Program.” World Development 129 (May): 104876. Heise, Lori L. 2001. What Works to Prevent Partner Violence: An Evidence Overview. London: Strive Research Consortium, London School of Hygiene and Tropical Medicine. Hidrobo, Melissa, and Lia Fernald. 2013. “Cash Transfers and Domestic Violence.” Journal of Health Economics 32 (1): 304–19. https://doi.org/10.1016/j.jhealeco.2012.11.002. Hidrobo, Melissa, Amber Peterman, and Lori L. Heise. 2016. “The Effect of Cash, Vouchers, and Food Transfers on Intimate Partner Violence: Evidence from a Randomized Experiment in Northern Ecuador.” American Economic Journal: Applied Economics 8 (3): 284–303. Hossain, Mazeda, Cathy Zimmerman, Ligia Kiss, Tanya Abramsky, Drissa Kone, Monika Bakayoko-Topolska, Jeannie Annan, Heidi Lehmann, and Charlotte Watts. 2014. “Working with Men to Prevent Intimate Partner Violence in a Conflict-Affected Setting: A Pilot Cluster Randomized Controlled Trial in Rural Côte d’Ivoire.” BMC Public Health 14 (April 10), article 339. Hughes, Christine, Mara Bolis, Rebecca Fries, and Stephanie Finigan. 2015. “Women’s Economic Inequality and Domestic Violence: Exploring the Links and Empowering Women.” Gender and Development 23 (2): 279–97. IASC (Inter-Agency Standing Committee). 2015. Guidelines for Integrating Gender-Based Violence Interventions in Humanitarian Action: Reducing Risk, Promoting Resilience, and Aiding Recovery. Geneva: IASC. https://gbvguidelines.org/wp/wp-content/uploads/2016/10/2015_IASC_Gender- based_Violence_Guidelines_full-res.pdf. Ismayilova, Leyla, Leyla Karimli, Eleni Gaveras, Alexice Tô-Camier, Jo Sanson, Josh Chaffin, and Rachel Nanema. 2018. “An Integrated Approach to Increasing Women’s Empowerment Status and Reducing Domestic Violence: Results of a Cluster-Randomized Controlled Trial in a West African Country.” Psychology of Violence 8 (4): 448–521. Abdul Latif Jameel Poverty Action Lab (J-PAL). 2021. “Designing financial services and social protection programs to enhance women’s economic empowerment.” J-PAL Policy Insights. Last modified February 2021. https://doi.org/10.31485/pi.3090.2021. Jensen, Robert, and Emily Oster. 2009. “The Power of TV: Cable Television and Women’s Status in India.” Quarterly Journal of Economics 124 (3): 1057–94. Jewkes, Rachel. 2002. “Intimate Partner Violence: Causes and Prevention.” Lancet 359 (9315): 1423–29. Jewkes, Rachel, Samantha Willan, Lori L. Heise, Laura Washington, Nwabisa Shai, Alice Kerr-Wilson, and Nicola Christofides. 2020. “Effective Design and Implementation Elements in Interventions to Prevent Violence against Women and Girls.” What Works to Prevent Violence Series, South African Medical Research Council, Pretoria. Johnson, Kiersten. 2003. “Dialectics of Power and Violence in the Home: A Comparative Analysis of Women’s Experience of Domestic Violence in Haiti and Nicaragua.” Dissertation, University of Maryland, College Park, MD. 82 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Kim, Julia C., Charlotte H. Watts, James R. Hargreaves, Luceth X. Ndhlovu, Godfrey Phetla, Linda A. Morison, Joanna Busza, John D. H. Porter, and Paul Pronyk. 2007. “Understanding the Impact of a Microfinance-Based Intervention on Women’s Empowerment and the Reduction of Intimate Partner Violence in South Africa.” American Journal of Public Health 97 (10): 1794–1802. Kishor, Sunita, and Kiersten Johnson. 2003. “Women’s Health at the Nexus of Poverty and Domestic Violence: Evidence from the Developing World.” Paper presented at the 2003 Annual Meeting of the Population Association of America, Minneapolis, MN, May 1–2, 2003. Kishor, Sunita, and Kiersten Johnson. 2004. Profiling Domestic Violence: A Multi-Country Case Study. June. Calverton, MD: ORC Macro. Kuttner, Stephanie, J. Ngo Likeng, K. Schuettler, and T. Yossa. 2020. “Étude qualitative sur les violences basées sur le genre et le Projet Filets Sociaux au Cameroun.” World Bank, Washington, DC. Kuttner, Stephanie, and Mpala Nkonkomalimba. 2020. “Safety for All: Gender Based Violence and Social Safety Nets in Zambia.” World Bank, Washington, DC. La Ferrara, Eliana, Alberto Chong, and Suzanne Duryea. 2012. “Soap Operas and Fertility: Evidence from Brazil.” American Economic Journal: Applied Economics 4 (4): 1–31. Lees, Shelley, Nambusi Kyegombe, Ampa Diatta, Amanda Zogrone, Shalini Roy, and Melissa Hidrobo. 2021. “Intimate Partner Relationships and Gender Norms in Mali: The Scope of Cash Transfers Targeted to Men to Reduce Intimate Partner Violence. Violence Against Women.” 27 (3–4): 447–69. Lindert, Kathy, Tina George Karippacheril, Inés Rodríguez Caillava, and Kenichi Nishikawa Chávez, eds. 2020. Sourcebook on the Foundations of Social Protection Delivery Systems. Washington, DC: World Bank. Litwin, Ashley, Elizaveta Perova, and Sarah Anne Reynolds. 2019. “A Conditional Cash Transfer and Women’s Empowerment: Does Bolsa Família Influence Intimate Partner Violence?” Social Science and Medicine 238 (October): 112462. https://doi.org/10.1016/j.socscimed.2019.112462. Mahmud, M., Orkin, K., and Riley, E. 2020. “Economic and Psychological Constraints to Women’s Empowerment.” Working Paper (July 9), University of Oxford, Oxford, UK. Palermo, Tia M., Jennifer Rebecca Bleck, and Amber Peterman. 2014. “Tip of the Iceberg: Reporting and Gender-Based Violence in Developing Countries.” American Journal of Epidemiology 179 (5): 602–12. Pathways of Women’s Empowerment. 2011. “Empowerment: A Journey Not a Destination.” Pathways of Women’s Empowerment, Institute of Development Studies, Brighton, UK. Pavanello, Sara, Carol Watson, W. Onyango-Ouma, and Paul Bukuluki. 2016. “Effects of Cash Transfers on Community Interactions: Emerging Evidence.” Journal of Development Studies 52 (8): 1147–61. Peterman, Amber, Anastasia Naomi Neijhoft, Sarah Cook, and Tia M. Palermo. 2017. “Understanding the Linkages between Social Safety Nets and Childhood Violence: A Review of the Evidence from Low- and Middle-Income Countries.” Health Policy and Planning 32 (7): 1049–71. Bibliography 83 Peterman, Amber, Benjamin Schwab, Shalini Roy, Melissa Hidrobo, and Daniel O. Gilligan. 2015. “Measuring Women’s Decisionmaking: Indicator Choice and Survey Design Experiments from Cash and Food Transfer Evaluations in Ecuador, Uganda, and Yemen.” IFPRI Discussion Paper 1453 (August), International Food Policy Research Institute, Washington, DC. Peterman, Amber, Else Valli, and Tia M. Palermo. 2021. “Government Anti-Poverty Programs and Intimate Partner Violence.” Economic Development and Cultural Change. Renner, Lynette M., and Kirsten Shook Slack. 2006. “Intimate Partner Violence and Child Maltreatment: Understanding Intra- and Intergenerational Connections.” Child Abuse and Neglect 30 (6): 599– 617. Riley, Emma. 2020. “Resisting Sharing Pressure in the Household Using Mobile Money: Experimental Evidence on Microenterprise Investment in Uganda.” Job Market Paper (November 9), Department of Economics, University of Oxford, Oxford, UK. Rivett, Mark, Emma Howarth, and Gordon Harold. 2006. “‘Watching from the Stairs’: Towards an Evidence- Based Practice in Work with Child Witnesses of Domestic Violence.” Clinical Child Psychology and Psychiatry 11 (1): 103–25. Rojas, Mariano. 2011. “Poverty and Psychological Distress in Latin America.” Journal of Economic Psychology 32 (2): 206–17. https://doi.org/10.1016/j.joep.2010.01.014. Rosas, Nina, and Shwetlena Sabarwal. 2016. “Can You Work It? Evidence on the Productive Potential of Public Works from a Youth Employment Program in Sierra Leone.” Policy Research Working Paper 7580, World Bank, Washington, DC. Roy, Shalini, Melissa Hidrobo, John Hoddinott, and Akhter Ahmed. 2019. “Transfers, Behavior Change Communication, and Intimate Partner Violence: Postprogram Evidence from Rural Bangladesh.” Review of Economics and Statistics 101 (5): 865–77. Sarma, Nayantara. 2020. “Domestic Violence and Workfare: An Evaluation of India’s MGNREGA.” Webinar, September 23, 2020. http://www.igidr.ac.in/seminars/webinar/. Schaner, Simone. 2017. “The Cost of Convenience? Transaction Costs, Bargaining Power, and Savings Account Use in Kenya.” Journal of Human Resources 52 (4): 919–45. Stets, Jan E. 1991. “Psychological Aggression in Dating Relationships: The Role of Interpersonal Control.” Journal of Family Violence 6 (1): 97–114. Subbarao, Kalanidhi, Carlo del Ninno, Colin Andrews, and Claudia Rodríguez-Alas. 2013. Public Works as a Safety Net: Design, Evidence, and Implementation. Directions in Development: Human Development Series. Washington, DC: World Bank. Tauchen, Helen V., Ann Dryden Witte, and Sharon K. Long. 1991. “Domestic Violence: A Nonrandom Affair.” International Economic Review 32 (2): 491–511. 84 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Usdin, Shereen, Esca Scheepers, Susan Goldstein, and Garth Japhet. 2005. “Achieving Social Change on Gender-Based Violence: A Report on the Impact Evaluation of Soul City’s Fourth Series.” Social Science and Medicine 61 (11): 2434–45. Vaillant, Julia, Estelle Koussoubé, Danielle Roth, Rachael Pierotti, Mazeda Hossain, and Kathryn L Falb. 2020. “Engaging Men to Transform Inequitable Gender Attitudes and Prevent Intimate Partner Violence: A Cluster Randomised Controlled Trial in North and South Kivu, Democratic Republic of Congo.” BMJ Global Health 5 (5): e002223. Valli, Else, Amber Peterman, and Melissa Hidrobo. 2019. “Economic Transfers and Social Cohesion in a Refugee Hosting Setting.” Journal of Development Studies 55 (Supplement 1): 128–46 Van Wyk, Judy A., Michael L. Benson, Greer Litton Fox, Alfred DeMaris. 2003. “Detangling Individual-, Partner-, and Community-Level Correlates of Partner Violence.” Crime and Delinquency 49 (3): 412–38. Vyas, Seema, and Charlotte Watts. 2009. “How Does Economic Empowerment Affect Women’s Risk of Intimate Partner Violence in Low and Middle Income Countries? A Systematic Review of Published Evidence.” Journal of International Development 21 (5): 577–602. https://doi.org/10.1002/jid.1500. Wasilkowska, Kamila. 2012. “Gender Impact Analysis: Unconditional Cash Transfers in South Central Somalia.” December 21, Somalia Cash Consortium, Nairobi, Kenya. WFP (World Food Programme). 2019. The Potential of Cash-Based Interventions to Promote Gender Equality and Women’s Empowerment: A Multi-country Study. Technical Report (February). Rome: WFP. WHO (World Health Organization). 2013. Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and nonpartner sexual violence. WHO (World Health Organization). 2016. Ethical and safety recommendations for intervention research on violence against women. https://www.who.int/reproductivehealth/publications/violence/ intervention-research-vaw/en/. WHO (World Health Organization). 2021. “Devastatingly Pervasive, 1 in 3 Women Globally Experience Violence: Younger Women among Those Most at Risk: WHO.” News Release, March 9, 2021. https://www.who.int/news/item/09-03-2021-devastatingly-pervasive-1-in-3-women-globally- experience-violence. WILPF (Women’s International League for Peace and Freedom). 2019. “Beyond Consultations: A Tool for Meaningfully Engaging with Women in Fragile and Conflict-Affected States.” March 14, WILPF, Geneva. https://www.peacewomen.org/node/103491. World Bank. 2014. Voice and Agency: Empowering Women and Girls for Shared Prosperity. Washington, DC: World Bank. World Bank. 2017. The World Bank Environmental and Social Framework. Washington, DC: World Bank. Bibliography 85 World Bank. 2018. “G20 Digital Identity Onboarding.” World Bank, Washington, DC. http://hdl.handle. net/10986/30484. World Bank. 2020a. “Childcare Goes Mobile in Burkina Faso.” Umbrella Facility for Gender Equality Results Series, World Bank, Washington, DC. World Bank. 2020b. “Environmental and Social Framework for IPF Operations: Addressing Sexual Exploitation and Abuse and Sexual Harassment (SEA/SH) in Investment Project Financing involving Major Civil Works.” Good Practice Note, 2nd ed. (February), World Bank, Washington, DC. World Bank. Forthcoming. “Addressing Sexual Exploitation and Abuse and Sexual Harassment: Good Practice, SEA/SH Risk Screening Tool.” Human Development Practice Group, World Bank, Washington, DC. World Bank. Forthcoming. “Codes of Conduct Brief.” World Bank, GWI (Global Women’s Institute), and IDB (Inter-American Development Bank). 2014. “Violence against Women and Girls (VAWG) Resource Guide: Introduction.” World Bank, Washington, DC. https://www.worldbank.org/content/dam/Worldbank/document/Gender/ VAWG%20Resource%20Guide%20Introduction%20July%202014.pdf. Yoong, Joanne, Lila Rabinovich, and Stephanie Diepeveen. 2012. “The Impact of Economic Resource Transfers to Women Versus Men: A Systematic Review.” Technical report 2001 (January), EPPI- Centre, Social Science Research Unit, Institute of Education, University of London, London. Yue, Zhiying, Hua Wang, and Arvind Singhal. 2019. “Using Television Drama as Entertainment-Education to Tackle Domestic Violence in China.” Journal of Development Communication 30 (1): 30–44. 86 SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE ANNEXES ANNEXES 87 88 Annex 1: Additional resources NAME LINKS DESCRIPTION CARE. Cash & Voucher Assistance https://gbvguidelines.org/en/documents/ The purpose of this cash and voucher assistance (CVA) and gender-based and Gender-Based Violence cash-voucher-assistance-and-gbv- violence (GBV) compendium is to assist humanitarian actors and crisis- and Compendium: Practical Guidance for compendium-practical/ conflict-affected communities to: Humanitarian Practitioners. 2019. 3 integrate GBV risk mitigation into CVA interventions; 3  integrate GBV prevention into multisectoral programming using CVA when appropriate; and 3 integrate CVA into GBV prevention and response when appropriate. The purpose of the Compendium is to assist humanitarian actors and crisis- and conflict-affected communities to: integrate Gender-Based Violence (GBV) risk mitigation and in some cases GBV prevention into Cash and Voucher Assistance (CVA) interventions; and integrate CVA into GBV prevention and response when appropriate. SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE The compendium is for: • Field-based humanitarian practitioners, across all areas or sectors of humanitarian response who use cash and/or vouchers in their programs; • GBV specialists who are considering using CVA in their programming; • Members of the humanitarian country team (HCT); • Humanitarian coordinators (HCs) and donors who advise and monitor teams and partners on GBV main- streaming/integration. The document helps practitioners to: • Differentiate between i) GBV risk mitigation in CVA, and ii) potential ways in which CVA can contribute to GBV prevention and response; • Identify practical actions that CVA/technical sector actors and other humanitarian actors can take to identify and mitigate the risks of GBV in CVA; • Identify practical actions that GBV specialists can take to incorporate GBV- protective CVA in their GBV programming, providing key considerations throughout the phases of project cycle management. IRC, Mercy Corps, Women’s Refugee https://www.womensrefugeecommission. • This toolkit assists practitioners in collecting the requisite situational Commission. no date. Mainstreaming org/issues/livelihoods/research-and- protection information on risks for affected populations with an age, gender, GBV Considerations in CBIs and resources/1549-mainstreaming-gbv- and diversity (AGD) lens, identifying community-based or self-protection Utilizing Cash in GBV Response: considerations-in-cbis-and-utilizing-cash- mechanisms, informing tailored and protective cash-based interventions, and Toolkit for Optimizing Cash-based in-gbv-response preparing a monitoring system that is based on identified protection risks. Interventions for Protection from • In Section I you will find a focus group discussion/interview tool and Gender-based Violence. accompanying guidance to assess and mitigate potential risks, as well as a post-distribution monitoring tool and accompanying guidance to monitor risks. • In Section II you will find a protocol to assess and address GBV survivors’ needs for cash assistance within GBV case management services, as well as a post-distribution monitoring tool and accompanying guidance to monitor risks. IASC. 2018. Gender Handbook for https://interagencystandingcommittee. A practical tool that provides guidance to frontline humanitarian professionals humanitarian action. (2018) org/iasc-reference-group-gender- on how to integrate gender equality and the empowerment of women and girls and-humanitarian-action/iasc-gender- throughout the humanitarian action cycle. handbook-humanitarian-action-2017 https://interagencystandingcommittee. org/gender-and-humanitarian-action/ content/gender-handbook-humanitarian- action Global Protection Cluster and IASC. https://gbvguidelines.org/en/ The “Pocket Guide” resource package V2.0 is a joint GBV Guidelines and GBV No date. How to support survivors of pocketguide/ AoR resource designed to provide all humanitarian practitioners with concrete gender-based violence when a GBV information on: actor is not available in your area: • How to support a survivor of gender-based violence who disclosed to you in A step-by-step Pocket Guide for a context where there is no gender-based violence actor (including a referral humanitarian practitioners’ version pathway or GBV focal point) available in your area. 2.0. The resource package uses global standards on providing basic support and information to survivors of GBV without doing further harm. We encourage adaptation of this resource to your local context with the support of a GBV specialist. WFP. 2016. Gender-based Violence https://gbvguidelines.org/en/documents/ This manual provides an overview of the key issues related to GBV in the Manual. world-food-programme-gender-based- context of WFP’s operations. It seeks to guide staff and partners so they can violence-manual/ better identify and respond to GBV risks related to hunger and nutrition and to WFP programs. ANNEXES 89 90 Danish Refugee Council, Global http://www.cashlearning.org/downloads/ This guide identifies minimum necessary information and key resources needed Protection Cluster, OXFAM, Save the erc-guide-for-protection-in-cash-based- to help humanitarian practitioners ensure that protection risks and benefits are Children, WFP, Women’s Refugee interventions-web.pdf considered and monitored throughout the CBI program cycle. Specifically, this Commission, UNHCR. No date. guide addresses: Guide for Protection in Cash-based Reducing risks: Interventions. Programs with CBI incorporate protective design, implementation and monitoring elements so that the program does not increase, and rather helps to mitigate, risks for beneficiaries or persons of concern. Design ensures that the introduction of cash does not exacerbate community tensions and relationships between beneficiaries and nonbeneficiaries of assistance, and monitoring of risks leads to program adjustments as necessary. Enhancing benefits: Design and revision of programs so that CBI enhances protection benefits such as improved household and community relations, dignity through choice, and safe, impartial access to assistance. Programs should build upon the inherent potential of CBI—a modality that enables the choice of affected people to use humanitarian aid as they see fit—to contribute to participation, accountability and meeting the needs of different groups and individuals. Reproductive Health Response in https://reliefweb.int/sites/ This manual is one of several outcomes of a three-year global Gender-based Conflict Consortium.2004. - Gender- reliefweb.int/files/resources/ Violence Initiative spearheaded by the Reproductive Health Response in Conflict based Violence Tools Manual For FC881A31BD55D2B3C1256F4F00461838- Consortium and aimed at improving international and local capacity to address SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE Assessment and Program Design, Gender_based_violence_rhrc_Feb_2004. gender-based violence (GBV) in refugee, internally displaced, and post-conflict Monitoring and Evaluation in pdf settings. The tools have been formulated according to a multisectoral model Conflict-Affected Settings. of GBV programming (described more thoroughly on page 35) that promotes action within and coordination between the constituent community, health and social services, and the legal and security sectors. The manual is meant to be used by humanitarian professionals who have experience with and are committed to GBV prevention and response. The tools are divided into three major categories: assessment, program design, and program M&E. Tools include: Tools Included in this Chapter • Situational Analysis Guidelines • Focus Group Guidelines • Community mapping assessment • Draft Prevalence Survey Questionnaire • Sample Interviewer Training Handbook • Assessment of existing multisectoral prevention and response • GBV staff recruitment guidelines • M&E tools (e.g. incident reporting, consent forms) World Bank. 2014. Violence Against https://openknowledge.worldbank.org/ This brief will specifically focus on four types of social protection interventions: Women and Girls: Social Protection handle/10986/21089?show=full&locale- social assistance, social insurance, labor market programs, and early childhood Brief. attribute=fr development. It will offer suggestions for integrating violence against women and girls prevention efforts within these interventions. These areas of focus are meant to be illustrative of different social protection programs, rather than to reflect the full breadth of SP programs. In general, SP programs are public interventions that support the poorest populations and assist individuals, households, and communities to better overcome social and economic risks. Examples of programs include: a) social assistance (SSNs): cash transfers, school feeding, and targeted food assistance; b) social insurance: old-age and disability pensions and unemployment insurance; c) labor market programs: skills-building programs, job-search and matching programs, and improved labor regulations; and d) early childhood development. Other program interventions, which fall under what is referred to as social protection, aim to strengthen families abilities to respond to hardships by promoting gender equality. Examples include early childhood development, projects that focus on at-risk youth, or targeted poverty alleviation programs. UNHCR. No date. Cash Assistance https://www.unhcr.org/protection/ This document outlines key considerations and learning to be used when and Gender - Key Considerations and operations/5bbf501b4/cash-assistance- planning to deliver and/or delivering cash assistance to refugees, internally Learning. gender-key-considerations-learning.html displaced persons, asylum-seekers, returnees and stateless people. It is primarily intended for UNHCR staff and partners and should be read together with other cash and protection related guidance. Global Protection Cluster, IASC. https://gbvguidelines.org/wp/wp- The purpose of these Guidelines is to assist humanitarian actors and No date. Guidelines for Integrating content/uploads/2016/01/2015-IASC- communities affected by armed conflict, natural disasters and other Gender-Based Violence Interventions GBV-Guidelines-main-book-without- humanitarian emergencies to coordinate, plan, implement, monitor and evaluate in Humanitarian Action. table-spreads.pdf essential actions for the prevention and mitigation of gender-based violence (GBV) across all sectors of humanitarian response. It provides specific guidance, organized into thirteen thematic area sections. Each section focuses on a different sector of humanitarian response. Although the guidance is organized in terms of discrete areas of humanitarian operation, all humanitarian actors must avoid ‘siloed’ interventions. ANNEXES 91 92 Annex 2: SSN project development objectives through a gender lens DEVELOPMENT GENDER DIMENSIONS, KEY OPPORTUNITIES RISKS SSN PROGRAM OPTIONS OBJECTIVE QUESTIONS, ESA Reduce poverty Is gender inequality a feature • Reach most vulnerable • Male backlash against • Targeting of gender-specific of poorest or most vulnerable women and close poverty- women transfer recipients subset of poor (for example, households? related gender gaps perceived as challenging widows, woman-headed men as breadwinners households) Are households with a woman • Reduce poverty-related primary income earner household stress • Community backlash against • Allow households to select (woman-headed households) • Reduce negative coping already marginalized women recipient or transfer to women poorer than households with strategies (sex-work, child • Inequitable distribution of • In polygamous households, male primary income earner labor, early marriage, and resources within households transfer to men, senior wives or (male-headed household)? so on) all wives • Opportunistic theft Are there intrahousehold SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE • Increase women’s economic or attacks on women • Set times / locations for pick- welfare inequalities? empowerment (if transfers beneficiaries as they travel up with women’s safety and Would the impact of the to women) to/from benefit pick-up transport options in mind intervention differ based on the sex of the designated recipient? Improve food security Who is responsible for purchase • Increase women’s agency in • If men traditionally As above (cash, food or voucher) of food and preparation of relation to food purchases responsible to purchase meals? (Are there different and intrahousehold food, may feel spending roles between men distribution. ‘disempowered’. and women?) • Increase access to food for • Gender or other inequity Is food distributed equitably women and children. in distribution of food between household members? resources Does this vary between • Possible conflict over different household structures? food distribution issues in polygamous or intergenerational households Improve livelihoods What are the main gender gaps • Promote women’s • Male backlash (spousal or • Targeting women for livelihoods in access to income generating economic empowerment other) against beneficiaries support activities (e.g. occupational • Nudge gender norm • Risks of harassment or • Engaging men to support segregation, wage gaps, access change regarding women’s assault while engaged in women’s engagement in to assets) engagement in productive livelihood activities productive activities What are the main constraints activities • Backlash against • Offer complementary life skills underlying these gender gaps? • Build women’s networks traditionally marginalized trainings to beneficiaries - (e.g. self-help groups) groups Gendered market analysis Improve human Are there gender gaps in • Close gender gaps in • Risks of reinforcing the gaps • Work with men and community development indicators health, education and nutrition human development (if giving transfer to male leaders to make program choices outcomes? outcomes or more senior women) or acceptable reinforcing women’s gender How are human capital • Increase access to services • Targeting women/girls only or norms re care of children investments made at the and potential to identify also boys/men household level? GBV and link to support • Risk of backlash if • ‘Labeling’ v ‘hard’ v ‘soft’ services prioritizing girls Who controls the resources conditions and makes decisions regarding • Secondary impacts of • SEA risks linked to using • Encourage engagement of men these investments? reducing future GBV risks services or verification and boys in domestic and care- of girls requirements giving labor. Provide wage-labor Will only women participate or • Increase women’s own • Risk of SEA/SH at or • Choice of nontraditional form of through public works will households select worker? income traveling to/from work sites labor What are the location, hours • Shift gender norms • Male backlash for • Set times/location taking and type and conditions of restricting ‘women’s work’ transgressing gender norms women’s security and domestic work? re ‘women’s work’ labor demands into account • Facilitate labor market How will women balance work integration • Male backlash for • Safety protocols and codes of requirements with domestic undermining husbands as conduct • Create gender-smart labor? primary breadwinners infrastructure to reduce • Oversight/inspection of work women’s labor and increase • Male or other household sites safety (e.g. accessible water member backlash for failing • Provisions for pregnant and sources, sex-segregated to complete expected lactating women public latrines, access to household duties childcare) • Facilitate childcare arrangements • Women only workers or mixed- gender teams • Choice of infrastructure to reduce women’s labor burden and increase women’s safety. Note: ESA = environmental and social assessment. ANNEXES 93 94 Annex 3: SEA/SH Risk Screening Tool indictors and risk mitigation measures TABLE 4. SEA/SH Risk Screening Tool Country Context Indicators (questions 1-13) INDICATOR RATIONALE FOR INDICATOR SCORING OPTIONS Country-level violence 4. Prevalence of intimate This statistic is intended to give an overview of levels of violence against women in the country. Higher Risk is having national IPV partner violence No direct correlation has been established between risk of SEA within a project and the levels of prevalence above regional average VAW in countries. However, this statistic is important for giving a sense of the country context per DHS within which the project occurs. The national IPV prevalence is compared with the regional Lower Risk is having national IPV average as per DHS regional estimates. prevalence below the regional average per DHS SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 5. Prevalence of any form of This indicator compares the national prevalence of sexual violence by any perpetrator with sexual violence the regional average. This statistic is intended to give an overview of levels of violence against women and girls in the country. 6. Prevalence of child This indicator compares the national prevalence of early/child marriage (marriage before the marriage exact age of 18) to global averages of child marriage and distributed into three categories of risk. This statistic captures another expression of violence experienced by women and girls. 7. State Department This indicator reflects the country’s classification according to the Trafficking in Persons Report. Trafficking in Persons This classification by Tier (I, II and III) is based on the extent of their governments’ efforts to Report comply with the “minimum standards for the elimination of trafficking.” 8. Is the country on the FCV This indicator captures whether the area where the project will/is being implemented is country list? undergoing a humanitarian or emergency crisis such as a natural disaster, conflict, epidemic or famine and is based on the World Bank’s most recent list of Fragile and Conflict-affected Situations. Legal context 9. Laws on sexual This indicator captures whether the country has laws on sexual harassment or not. Recognizing harassment that sexual harassment is a serious issue with repercussions for a country’s ability to welcome women into the workforce is a first step to creating a more equitable environment. 10. Lawsonmaritalrape This indicator captures whether the country has laws banning marital rape. Marital rape occurs when there is sexual intercourse between spouses without one party’s consent. It is a form of domestic violence and a violation of a woman’s human right to decide whether and when to have sexual relations. 11. Laws on domestic This indicator captures whether the country has laws banning domestic violence. Domestic violence violence may take the form of emotional or psychological abuse, physical abuse, or sexual abuse and has negative consequences for the mental, physical, and reproductive health of the victim, as well as potentially for those also living in the abusive household. Gender norms and beliefs 12. Justification of wife This indicator compares the percentage of women who agree that a husband is justified in beating hitting or beating his wife for at least one specific reason compared to the regional average. Help seeking to stop 13.  This indicator compares the percentage of ever-married women who have ever experienced any violence physical or sexual violence who sought help to stop violence either by telling someone or by seeking support from services with the regional average. Among women who seek help, most look to family and friends and very few look to formal institutions and mechanisms. National-level capacity to respond to GBV Existence of a national 14.  This indicator captures whether the country has a National Action Plan on addressing violence action plan on Women, against women and girls/GBV which may indicate the country’s commitment to addressing Peace and Security pressing GBV issues. The plans are a good source for sectoral commitments and priorities for addressing GBV. 15. GBV working group This indicator captures whether the country has a national gender-based violence working group. GBV Working groups seek to ensure more effective, accountable and comprehensive GBV prevention and response. This engagement suggests a commitment to addressing pressing GBV issues in the country. 16. National referral This indicator captures whether the country has a national referral pathway protocol on GBV. A pathway protocol on referral system helps women, girls, other at risk-groups and GBV survivors to safely and quickly GBV access health, psychosocial, protection, legal and socio-economic services and support. ANNEXES 95 96 TABLE 5. SEA/SH Risk Screening Tool project-related indicators(questions 14-25) ASSESSMENT QUESTION RATIONALE AND EXPLANATION SCORING OPTIONS Project Implementation Areas  Are project activities 17. Humanitarian emergencies may be man-made (internal or external conflict) or the Higher Risk [0] is when no project activities implemented in areas of result of climate shocks (earthquakes, floods, and so on). They typically result in are implemented in areas experiencing a the country experiencing a acute dependence on assistance and often in a significant influx of external actors humanitarian humanitarian emergency? (relief workers, security forces, and so on). During these humanitarian emergencies, Lower Risk [2] is when any project activities women and girls’ vulnerability to SEA/H can be increased due to acute dependence are implemented in areas of the country on project actors for emergency assistance and increased difficulties in monitoring experiencing a humanitarian emergency project-related SEA/SH risks. During displacement, unaccompanied women and children or those living in camps might also be particularly vulnerable to SEA. 18. Are project activities This question refers to systems established by the Borrower, whether financed by Low risk [0] is when the Borrower’s implemented in areas where the project or from other sources. The Borrower’s monitoring capacity might be monitoring capacity is sufficient the implementing agency’s limited because of capacity constraints (at either national or local levels) and/or High risk [1] is when activities are capacity to monitor the accessibility constraints due to insecurity or logistical challenges (difficult terrain, implemented in areas where the Borrower’s project is limited? distance, and so on). monitoring capacity is limited Project Preparation Process  s part of the project 19. A Consultations with local women’s groups, groups that advocate for children and Low risk [0] is when there were stakeholder SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE preparation, was there adolescent rights, women’s leaders and other stakeholders can help to understand consultations with groups advocating for meaningful consultation the local gender and GBV dynamics within which the project will be implemented. women, children and adolescent girls held with groups advocating This in turn can help identify potential project-related SEA/SH risks. In contexts separately from men for women, children and where the ability of women and girls to express their needs and concerns may be High risk [1] is when there were no adolescent girls? limited, effective consultation requires providing the opportunity for women to stakeholder consultations with groups participate separately or in women-only groups.99 The risk tool is expected to be advocating for women, children and filled out at both Concept Stage and at Appraisal. Risk is likely to be initially high adolescent girls held separately from men [i.e. unless consultations were carried out early on during the identification process] and low once the SEP and ESA have been implemented. 20. During stakeholder While concerns about SEA/SH may not arise during stakeholder consultations for Low risk [0] is when concerns related to consultations, did groups various reasons (because they are not a concern or, conversely, because SEA/SH is project-related SEA/SH were not raised advocating for women, too taboo to discuss), the fact that such concerns are raised can indicate that the High risk [1] is when concerns related to children and adolescent project may create additional SEA/SH risks. project-related SEA/SH risks were raised or girls raise concerns about Moderators should be able to respond if issues of SEA/SH or GBV more generally when there were no effective consultations the project’s potential arise, to maintain the privacy, dignity and ensure the well-being of all participants. with groups advocating for women, children additional SEA/SH risks? (Under no circumstances should participants be asked directly to reveal their and adolescent girls. personal experiences of GBV or reveal the identities of others having experienced GBV). 99 Beyond Consultations: A Tool For Meaningfully Engaging with Women in Fragile and Conflict-Affected States https://www.peacewomen.org/node/103491 Intervention Design 21. Do mechanisms for the When project actors100 have influence over the selection of beneficiaries, they may Low risk [0] is when individual project selection of beneficiaries abuse their position of power to extract sexual favors. Automated, randomized or actors have limited or no decision- making create opportunities for centralized beneficiary selection systems (such as those using social registries) to power over beneficiary selection. individual project actors to identify eligible households can significantly reduce these risks. In addition selection Medium risk [1] is when individual project sexually exploit or abuse mechanisms can include the presence of community members or authorities during actors have some degree of decision making beneficiaries? the process, that would minimize the potential for SEA/H. Projects where beneficiary power over beneficiary selection, but the selection is highly dependent on individual project actors represent a higher level of process takes place in public or in the risk. presence of community members or local authorities. High risk [2] is when individual project actors select beneficiaries with limited presence of external actors to validate the process. 22. Do mechanisms for During the process of verification of beneficiaries’ compliance with various Low risk [0] is when project actors have no the verification of conditions (e.g. utilizing prenatal services, regularly attending school, registering decision-making power over verification of conditionalities create with the national ID system), the project actors responsible for validating compliance conditionalities or there is effective oversight opportunities for project could seek sexual favors. Transparent procedures can include automatic use of Medium [1] when the verification of actors to sexually exploit or administrative data (e.g. school presence systems) or technological solutions (e.g. conditionalities is conducted by individual abuse project beneficiaries? thumbprint readers). In addition, community or public validation mechanisms can be project actors in public or in the presence of introduced expected to limit opportunities for SEA/H. This can include the presence community members/local authorities. of community members, multiple service providers, or authorities during the process High risk [2] is when the verification of of validation of compliance. Where the validation of conditionalities is reliant on conditionalities is conducted by individual assessments done by individual project actors with individual beneficiaries in private project actors in private settings. setting/without a community or public validation process the risk is expected to be higher. 23. Do mechanisms for During the transfer of benefits, project actors may abuse their position of power to Low risk [0] is when individual project the transfer of benefits sexually abuse or exploit beneficiaries. When benefits are transferred automatically, actors have no direct control over transfer of (cash, vouchers, in-kind the risk is considered low. Where project actors carry out transfers in the presence of benefits goods, stipends, wages, other individuals at the time of in-person distribution it is expected that the risk will Medium risk [1] when the transfer of and scholarships) create be reduced. Such individuals include but are not limited to community members or benefits is conducted by individual project opportunities for project local authorities. Transfer of benefits that occur through individual interactions with actors in public or in the presence of actors to sexually exploit or project actors in private settings which would pose a higher risk of SEA/H. community members or local authorities. abuse project beneficiaries? High risk [2] is when project actors have direct control over transfer of benefits and transfers occur in private settings. 100 As per the ESF, project actors include actors directly involved in its implementation, such as: (a) people employed or engaged directly by the Borrower to work specifically in relation to the project; (b) people employed or engaged through third parties to perform work related to core functions of the project; (c) people employed or engaged by the Borrower’s primary suppliers; and (d) people employed or engaged in providing community labor such as voluntary services or participation in project activities and processes. ANNEXES 97 98 24. Do project activities In many social protection and jobs programs, beneficiaries participate in group Low risk [0] is when there are no such include regular interaction activities such as public works, training courses, social care activities, social activities. with project actors (e.g. mobilization or information meetings, and so on. They may also be required to High risk [1.5] is when there are such participation in public participate in one-on-one activities such as counselling, mentoring, on-the-job opportunities during regular interaction works, attending regular training, or employment. Beneficiaries may also be at risk of sexual abuse or between beneficiaries and project actors. information or training harassment by other beneficiaries during group activities. Oversight of these Very high risk [2] is when there is a sessions, counselling), that activities may be ensured by the presence of other project actors, beneficiaries, residential component to the activity (such could create opportunities community members, or non-project actors. as boarding schools). for project actors to sexually exploit or abuse beneficiaries (or for sexual exploitation or abuse between beneficiaries)? 25. During program Female project actors (direct employees, service providers or contractors) may be Low risk [0] is when female project actors implementation, do female at risk of SH by male project actors, including managers and co-workers, particularly do not work with male actors alone or there project actors work with when working alone with them. (See note for question 18 on effective oversight.) is oversight of the interactions male project actors alone High risk [1] is when female project actors or with limited oversight? work with male project actors alone with no or limited oversight SAFETY FIRST: SOCIAL SAFETY NETS AND GENDER-BASED VIOLENCE 26. Does the program involve Some programs rely on police, peacekeepers, military personnel or armed local Low risk [0] is when there is no direct engagement with military militias for security. If in direct contact with beneficiaries, these forces may contact between military or paid security or paid security forces who perpetrate SEA and may not be under the jurisdiction of the national legal system. forces and beneficiaries or there is oversight come in direct contact High risk [1] is when there is direct contact with beneficiaries? between military or paid security forces and beneficiaries with limited oversight 98 Project Management 27. Does the implementing Codes of conduct establish what constitutes unacceptable behavior of SEA/SH for Low risk [0] is when codes of conduct agency already have all project actors, as well as the consequence of non-compliance. Their existence explicitly prohibiting SEA/SH do exist established codes may serve as a deterrent to SEA/SH by project actors. Codes of conduct include High risk [1] is when codes of conduct of conduct explicitly definitions of SEA/SH, project actors’ responsibilities, reporting protocols, sanctions, prohibiting SEA/SH do not exist prohibiting SEA/SH that and so on. would apply to project actors? 28. Have these codes of For codes of conduct to be an effective deterrent, they should be communicated to Low risk [0] is when codes of conduct conduct explicitly all project actors (including volunteers) and clearly define SEA/SH and lay out the explicitly prohibiting [SEA/SH] risks have prohibiting SEA/SH been consequences of non-compliance. been communicated to project actors. communicated to project High risk [1] is when codes of conduct actors? explicitly prohibiting [SEA/SH] risks have not been communicated to project actors or where they do not exist. ANNEXES 99