Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa This policy brief presents an update on the state of domestic violence or other violence against women in the Middle East and North Africa (MENA) region during June 2023 the COVID-19 pandemic using findings from a large-scale phone survey in 10 MENA countries, conducted by the World Values Survey (WVS) Association. The survey interviewed 12,366 respondents, distributed almost equally across Algeria, Palestine, Iran, Iraq, Jordan, Lebanon, Morocco, Saudi Arabia, Tunisia, and Egypt. Middle East Key Findings and North 1) In all countries, more than 63 percent of individuals reported an increase in violence or other abuse towards women in their community from their Africa Gender husbands or other family members 2) Women have a higher perception of domestic violence than men Innovation Lab 3) There is a concern about the presence of underreporting issues and a lack of The Middle East and North Africa awareness regarding the definition and recognition of domestic violence or abuse Gender Innovation Lab (MNAGIL) 4) More equitable gender attitudes and higher education are associated with carries out impact evaluations and a lower prevalence of domestic violence in the community and physical inferential research to produce arguments among family members rigorous evidence on what works in closing gaps in assets, economic opportunities, and agency between Context women and men, and how closing Domestic violence against women in the MENA region is a pressing and these gaps can help achieve other persistent issue. According to projections by the United Nations (UN) Women, development outcomes. Ultimately, approximately 37 percent of women in the Arab region have encountered some MNAGIL promotes the adoption of form of violence throughout their lives (slightly higher than the global average of policies and interventions that are one in three women). Looking at a broader definition of gender-based violence, shown to close gender gaps and such as child marriage and female genital mutilation, there are indications improve the well-being of women that this percentage could be even higher.1 Measures implemented to control and men in the Middle East and the spread of the COVID-19 pandemic, such as lockdowns and movement North Africa. restrictions, confined survivors and perpetrators in the same physical space. Further, stress due to economic uncertainty could trigger conflict within the family that can exacerbate domestic violence The Due to the difficulty of and ethical concerns around collecting data on gender- based violence (GBV), measures of prevalence and incidence of GBV are generally scarce across the globe, but especially in the MENA region. GBV surveys are typically conducted face-to-face to ensure the privacy, confidentiality, and 1 UN Women – Arab States. 2022. Facts and Figures: Ending Violence against Women and Girls. #2 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa security of respondents. Mobility constraints during COVID-19 render phone surveys the only feasible option to collect data. To ensure the safety of respondents, World Bank researchers in consultation with GBV experts, develop proxy questions asking perception of violence in the community, instead of querying individual experiences of violence.2,3 The same approach is later adopted by UN Women’s Rapid Gender Assessment surveys conducted in 13 countries, primarily outside of the MENA region. The report finds 45 percent of women reported that they or someone they know has experienced some form of violence against women since the onset of the pandemic.4 It is also adopted by the World Values Survey used in this brief. What did we do?234 We use a novel survey collected in collaboration with the (69 percent), Algeria (63 percent), Jordan (57 percent), Egypt World Values Survey (WVS) amidst the COVID-19 pandemic (56 percent), Palestine (56 percent), and Iran (53  percent). in 2021 to analyze the gender differences in perceptions This surge can be attributed to multiple factors, including of domestic violence by women and men. World Values increased stress and anxiety, enforced isolation and Surveys are cross-sectional surveys that assess individual confinement measures, disruption of supportive networks, attitudes and values, such as cultural values; attitudes and as well as intensified power dynamics and control within beliefs towards gender, family, and religion; attitudes and households. Even in high-income countries such as Saudi experience of poverty; education, health, and security; social Arabia, where people generally faced fewer socio-economic tolerance and trust; and attitudes towards multilateral stressors during the pandemic, such as unemployment institutions. In addition to the standard WVS modules, and other stressors such as food insecurity, a significant this round includes questions related to COVID-19, such 38 percent of respondents reported an increase in domestic as perceptions of COVID-19 effects on the respondent, violence or other abuse towards women since the beginning their family, and in the community. The survey covers 10 of the pandemic. It is worth noting, however, that respondents countries in the MENA region: Algeria, Palestine, Iran, Iraq, in Saudi Arabia reported the lowest increase in domestic Jordan, Lebanon, Morocco, Saudi Arabia, Tunisia, and violence against women compared to other countries. This Egypt. A total of 12,366 adult respondents (aged 18+) were could highlight the potential impact of economic stressors on interviewed via phone between March and June 2021.5 The domestic violence against women. survey is representative at the national level.6 What did we find? The impact of the COVID-19 The impact of the COVID-19 pandemic on domestic violence against women in MENA communities was devastating, with pandemic on domestic violence countries experiencing fewer economic stressors being no against women in MENA exception. More than 63 percent of respondents in 10 MENA communities was devastating, countries reported, on average, an increase within their community in violence or other abuse towards women from with countries experiencing their husbands or other family members since the beginning fewer economic stressors of the COVID-19 pandemic in March 2020. The largest being no exception. increase was reported in Lebanon (82 percent of respondents), followed by Tunisia (80 percent), Morocco (78 percent), Iraq 2 Halim, Daniel, England Rhys Can, and Elizaveta Perova. 2020. What Factors Exacerbate and Mitigate the Risk of Gender-Based Violence During COVID-19? 3 Perova, Elizaveta; Jarvis, Forest Brach; Arango, Diana Jimena; Peterman, Amber. Can We Capture Exposure to Gender-based Violence (GBV) through Phone Surveys during a Pandemic? 4 UN Women. 2022. Measuring the Shadow Pandemic: Violence against Women during COVID-19. 5 There are slight differences in the interview timeframe. For example, in Lebanon, the survey was concluded in a week (April 2-8), while it took two months to finish the survey in Iran (April 14-June 19). All 10 country surveys started in a similar timeframe, between March 28 and April 17. 6 Stratified random sampling was used to minimize selection bias. However, it is worth noting that the response rate varies across coun- tries, from a low of 8 percent in Iran to a high of 79 percent in Lebanon and Morocco. #3 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Expectedly, female respondents tend to have a higher of the challenges faced by women in their communities. perception of domestic violence compared to male The variation in perceptions based on the gender of the respondents. The most significant gender differences in respondent is not as high as anticipated though. responses are observed in Saudi Arabia, Egypt, and Algeria (Figure 1). This is because women are more directly affected Expectedly, female respondents tend to by and have a deeper understanding of the issue due to their personal experiences, societal roles, and awareness have a higher perception of domestic violence compared to male respondents. Figure 1 Share of individuals reporting within their community in domestic violence against women since the beginning of COVID-19 100 90 80 70 60 50 40 30 20 Male 10 0 Female Algeria Egypt Iran Iraq Jordan Lebanon Morocco Palestine Saudi Arabia Tunisia Note: 95% Confidence Interval Source: Author’s calculations using data from the World Values Survey, World Bank, 2021. The substantial increase in domestic violence or abuse Our analysis also reveals considerable heterogeneity across against women reported at the community level in MENA countries in the reported occurrence of physical arguments at during the COVID-19 pandemic does not align with the the household level (Figure 2A). The percentage of respondents reported physical arguments at the family level, suggesting reporting occasional or frequent physical arguments within concerning underreporting issues or a lack of awareness their families ranges from as high as 41 percent in Morocco regarding the definition and recognition of domestic violence and Algeria, to as low as 7 percent in Iran. The percentage falls or abuse. The WVS has specific questions that differentiate within moderate levels in the remaining countries: Tunisia between observations of domestic violence or abuse against (28 percent), Lebanon (24 percent), Egypt (23 percent), Iraq women at the community level and personal experiences of (21 percent), Palestine (16 percent), Jordan (15 percent), and physical and verbal arguments within families. While over 63 Saudi Arabia (15 percent). percent of respondents reported an increase in violence or other abuse towards women in their community since the beginning of COVID-19, only 23 percent of respondents in our The substantial increase in domestic sample reported occasional or frequent physical arguments violence or abuse against women within their families in the 10 months prior to the survey. It is not entirely surprising to observe a discrepancy between what reported at the community level in MENA respondents report about the situation in their community during the COVID-19 pandemic does and what is happening within their own households. While not align with the reported physical lack of awareness may contribute to this pattern, it is likely that individuals feel safer to reveal truthfully the prevalence arguments at the family level, suggesting of gender-based violence at the community level, even concerning underreporting issues or extrapolating from their own household to the community. a lack of awareness regarding the However, due to social norms, shame, fear of retaliation, and other factors, they are less likely to report arguments definition and recognition of domestic occurring within their own household. violence or abuse. #4 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Figure 2 Panel A. Share of individuals reporting there were physical arguments between their family Physical members sometimes or often in the 10 months preceding the survey 100 Arguments 90 80 70 60 50 40 30 20 Male 10 0 Female Algeria Egypt Iran Iraq Jordan Lebanon Morocco Palestine Saudi Arabia Tunisia Note: 95% Confidence Interval Source: Author’s calculations using data from the World Values Survey, World Bank, 2021. Panel B. Share of individuals reporting there were physical arguments between their family Verbal members sometimes or often in the 10 months preceding the survey 100 Arguments 90 80 70 60 50 40 30 20 Male 10 0 Female Algeria Egypt Iran Iraq Jordan Lebanon Morocco Palestine Saudi Arabia Tunisia Note: 95% Confidence Interval Source: Author’s calculations using data from the World Values Survey, World Bank, 2021. Interestingly, in half of the 10 MENA countries, a larger violence or other forms of abuse specifically directed percentage of male respondents (compared to female toward women. This could be because reporting violence or respondents) reported occasional or frequent physical abuse towards women may implicate men as perpetrators arguments (Figure 2A). In Algeria, for example, 43 percent of or raise questions about their behavior and actions. Further male respondents reported occasional or frequent physical research is needed, however, to delve deeper into the arguments compared to 40 percent of female respondents. underlying reasons for this observed difference. The respective percentages are also larger for males than females in Morocco (42 percent, 41 percent), Lebanon (25 percent, 23 percent), Iraq (22 percent, 20 percent), and Interestingly, in half of the 10 MENA Iran (8 percent, 7 percent). This pattern cannot be fully explained by differences in reporting behavior because we countries, a larger percentage of male do not observe the same pattern when reporting on domestic respondents (compared to female violence at the community level (Figure 1). It is possible respondents) reported occasional or that men may feel more comfortable or socially accepted in acknowledging and reporting physical arguments within frequent physical arguments their families compared to acknowledging and reporting #5 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Photo ©: Flickr / Matthew Lowe The COVID-19 pandemic was additionally associated with Socioeconomic shocks caused by the COVID-19 increased verbal arguments between family members in MENA. pandemic—such as a death in the family, job and income More than half of respondents (51 percent) in the 10 MENA loss—are associated with an increased likelihood of countries of our sample reported occasional or frequent domestic violence, physical and verbal arguments. verbal arguments within their families in the 10 months prior Regression estimates suggest that economic stressors to the survey. The largest increase was reported in Morocco following the COVID-19 pandemic, such as death/illness (71 percent of respondents), followed by Lebanon (65 percent in the family, the main income earner losing their job, of respondents), Tunisia (61 percent), Algeria (61 percent), Iran and a decline in household income are associated with (46 percent), Egypt (45 percent), Jordan (41 percent), Palestine a higher likelihood of reporting domestic violence in the (42 percent), Saudi Arabia (42 percent), and Iraq (42 percent). community, as well as physical and verbal arguments While verbal arguments between family members alone may between family members (Figure 3).7 This is true for both not always be considered domestic violence, they can be a male and female respondents, and after controlling for potential indicator or precursor of an unhealthy or abusive demographic characteristics and contextual differences relationship. Importantly, there are no significant differences across countries. between male and female respondents in reporting verbal arguments between family members (Figure 2B). This suggests that both genders are equally likely to acknowledge Socioeconomic shocks caused by the and report such conflicts within their families. COVID-19 pandemic—such as a death in the family, job and income loss— The COVID-19 pandemic was additionally are associated with an increased associated with increased verbal arguments likelihood of domestic violence, between family members in MENA. physical and verbal arguments. 7 The OLS regressions control for COVID-19-related shocks (death/illness in the family, main income earner losing their job, and decline in household income), an index for gender-equitable values, educational attainment, respondent’s age, marital status, and urban residence, as well as country fixed effects. Regression estimates of verbal arguments are not shown, but all the coefficients are pointing in the same direction. #6 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Figure 3 Share of individuals reporting there were physical Share of individuals reporting an increase arguments between family sometimes or often in in domestic violence against women in their the 10 months preceding the survey (%) community since the beginning of COVID-19 (%) Panel A Panel B Physical arguments Domestic violence in the community 10 12 5 8 0 4 -5 0 -10 -4 Male -15 -8 Female Family Main income Household Gender- Secondary- Family Main income Household Gender- Secondary- member earner lost income equitable educated member earner lost income equitable educated ill/died their job declined values or above ill/died their job declined values or above Note: the chart shows coefficients from an OLS regression of physical arguments among family members (Panel A) and domestic in the community (Panel B) on economic stressors since the COVID-19 pandemic (death/illness in the family, main income earner losing their job, a decline in household income), an index of gender-equitable values, a dummy for having at least secondary education, demographic characteristics (age, urban residence, and marital status), and country fixed effects. The regressions are estimated separately for male and female samples. Positive (negative) coefficients indicate an increase (decrease) in the likelihood of physical arguments and domestic violence. Solid-colored bars indicate statistically significant coefficients and light-colored bars indicate not statistically significant coefficients. Source: Authors’ calculations using World Values Survey data (World Bank, 2021). Education and positive attitudes toward gender equality Education and positive attitudes toward gender are powerful mitigating factors. Men and women with more gender-equitable values and with at least secondary equality are powerful mitigating factors. education are less likely to report domestic violence in the community and physical arguments among family members.8 In the case of physical arguments, these two factors completely offset the increased likelihood due to all economic stressors combined—for both women and men (Figure 3). The mitigating effects of education and gender-equitable values are not as pronounced on—but still largely point towards—lowering the likelihood of domestic violence in the community.9 Women’s education may have opposite effects on the reporting of domestic violence in the community. On one hand, education may empower women and improve their bargaining power, which reduces their likelihood of experiencing domestic violence or abuse. On the other hand, education may also improve women’s awareness of what is considered domestic violence. Further Photo ©: Shutterstock / ZouZou research is needed to disentangle these two effects. 8 The gender-equitable values index takes a simple average of 11 ordinal questions on attitudes and values towards equality between men and women. All 11 questions have a Likert scale (1-4) ranging from “Strongly Disagree” to “Strongly Agree”. We reverse the order on some of the questions to ensure that larger values consistently point to more equal gender values and attitudes. Please refer to Chaudhury, El-Shal, and Halim (2023) for more details on its construction and the list of 11 questions. 9 Except for secondary education among women which is not a statistically significant factor. Gender-equitable values for men are also not a statistically significant factor. #7 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Our results suggest a few policy implications: 1. Data is needed to measure the prevalence and incidence of gender-based violence (GBV). Better data sheds light on the issue and enables policymakers to prepare policy responses to safeguard and protect survivors. Data collection on gender-based violence should adhere to “safety first” principles, taking extra precautions to ensure the safety, security, and privacy of respondents. Data should be collected with the intention of reducing the incidence of violence and better addressing the needs of GBV survivors, instead of collecting data for data’s sake. Photo ©: Shutterstock / Dapetrus 2. By providing education and promoting Insights for policy women’s empowerment and positive attitudes towards gender equality, it is possible to Persistent domestic violence in the MENA region, which mitigate the occurrence of domestic violence has been further exacerbated by the COVID-19 pandemic, and alleviate its well-being impact on women will have profound gendered implications on the well- across the MENA region. being of women. Addressing domestic violence is crucial not only for safeguarding women’s well-being and the well-being of their families and communities but also for promoting economic growth and development, as women 3. Efforts to address domestic violence must thrive and contribute to their full potential. Globally, the include a focus on transforming gender overall direct and indirect costs associated with violence attitudes and norms. Promoting respect, against women are estimated to range from 1 percent equality, and non-violence in relationships is to 2 percent of the Gross National Product (GNP).10 This a core strategy for creating lasting change. translates to substantial economic costs at the regional level in MENA, amounting to millions of dollars. In Egypt, for example, it was estimated in 2015 that marital violence leads to the loss of more than 500,000 working days 4. Policymakers in MENA should pay attention annually, and the health sector incurs a cost of over USD to the underreporting of domestic violence 14 million to provide services to only a quarter (600,000) and develop contextually-relevant strategies of the survivors.11 to overcome the barriers preventing survivors from seeking help while simultaneously addressing the underlying factors contributing Persistent domestic violence in the to underreporting. MENA region, which has been further exacerbated by the COVID-19 pandemic, will have profound gendered implications on the well-being of women. 10 UN Women – Arab States. 2022. Facts and Figures: Ending Violence against Women and Girls. 11 Central Agency for Public Mobilization and Statistics, National Council for Women in Egypt, UNFPA. 2015. The Egypt Economic Cost of Gender-Based Violence Survey. #8 Action Alert? Endemic Gender-Based Violence in the Middle East and North Africa Acknowledgments This brief is a product of MNAGIL. It was prepared by Nazmul Chaudhury, Amira El-Shal, and Daniel Halim. We gratefully acknowledge funding from the Umbrella Facility for Gender Equality (UFGE) to carry out this work. UFGE has received generous contributions from Australia, Bill & Melinda Gates Foundation, Canada, Denmark, Finland, Germany, Iceland, Latvia, Netherlands, Norway, Spain, Sweden, Switzerland, United Kingdom, and the United States. For more information Nazmul Chaudhury nchaudhury@worldbank.org Daniel Halim dhalim@worldbank.org References Central Agency for Public Mobilization and Statistics, National Council for Women in Egypt, UNFPA. 2015. The Egypt Economic Cost of Gender-Based Violence Survey. [online] Available at: https://egypt.unfpa.org/en/publications/economic-cost-gender-based-violence- survey-egypt-2015 Chaudhury, Nazmul; El-Shal, Amira; Halim, Daniel. 2023. “Gender and Food Security in MENA.” Policy Research Working Papers, Forthcoming. Washington, DC: World Bank. Halim, Daniel, England Rhys Can, and Elizaveta Perova. 2020. What Factors Exacerbate and Mitigate the Risk of Gender-Based Violence During COVID-19?. East Asia & Pacific Gender Innovation Lab Policy Brief. Washington, DC: World Bank. Perova, Elizaveta; Jarvis, Forest Brach; Arango, Diana Jimena; Peterman, Amber. Can We Capture Exposure to Gender-based Violence (GBV) through Phone Surveys during a Pandemic?. Measurement Note. Washington, D.C. : World Bank Group. UN Women – Arab States. 2022. Facts and Figures: Ending Violence against Women and Girls. [online] Available at: https://arabstates.unwomen.org/en/what-we-do/ending-violence-against-women/ facts-and-figures-0 UN Women. 2022. Measuring the Shadow Pandemic: Violence against Women during COVID-19. New York: UN Women; 2021. Available at: https://data.unwomen.org/publications/vaw-rga