REPUBLIC OF SOUTH AFRICA SYSTEMATIC COUNTRY DIAGNOSTIC BACKGROUND NOTE: CRIME, VIOLENCE, & EXCLUSION IN SOUTH AFRICA SOCIETY OVERCOMING THE LEGACY OF EXCLUSION IN SOUTH AFRICA Alys Willman, Chandré Gould, Gareth Newham and Margarita Puerto Gomez Republic of South Africa Systematic Country Diagnostic | Acknowledgements ACKNOWLEDGEMENTS This paper was coauthored by Alys Willman, Senior Consultant, World Bank; Margarita Puerto Gomez, Senior Social Development Specialist, World Bank; Chandré Gould, Senior Researcher, Institute for Security Studies; and Gareth Newham, Head of Justice and Violence Prevention, Institute for Security Studies. Leonie Willenbrink, and Hope Lotsha Muronga, consultants, World Bank, also provided input. The paper benefitted from the very useful comments of Melanie Dugmore, Independent Consultant, GIZ; Terrence Smith, Program Manager, Violence Prevention Program, GIZ; and World Bank colleagues Stavros George Stavrou, Senior Social Development Specialist; Yaa Oppong, Senior Social Development Specialist, Marek Hanusch, Senior Economist and Erwin De Nys, Program Leader. i Contents | Republic of South Africa Systematic Country Diagnostic CONTENTS 1 INTRODUCTION 1 2 CRIME & VIOLENCE: KEY CONCEPTS 3 3 VIOLENCE IN SOUTH AFRICA: DATA, DEFINITIONS, & KEY TRENDS 5 3.1. Homicide 6 3.2. Violence Against Women 7 3.3. Violence Against Children 9 3.4. Youth Violence 9 3.5. Non-violent Crimes 10 3.6. Crimes Against Businesses 12 3.7. Perceptions of Insecurity and their Impact on Behavior 12 4 SPATIAL DISTRIBUTION OF CRIME & VIOLENCE 14 5 RISK FACTORS FOR VICTIMIZATION & PERPETRATION OF VIOLENCE & CRIME IN SOUTH AFRICA 16 5.1. Individual Level 18 5.2. Relationship Level 20 5.3. Community Level 21 5.4. Societal Level 22 6 THE STATE’S RESPONSE: AN OVERVIEW 25 6.1. Key Legislation 25 6.2. Key Institutions, Mandates, and Policies 28 6.3. Spotlight on the Criminal Justice Sector 32 7 CONCLUSION 35 APPENDIX A. DEFINING TYPES OF VIOLENCE 37 REFERENCES 38 i ii 1 INTRODUCTION This background note has been prepared as part of the Systematic Country Diagnostic (SCD) to guide the World Bank’s operations in South Africa. It provides an overview of the current situation, with particular attention to its roots in the country’s history of inequality and exclusion. For many South Africans, violence is a part of daily life. It exists in their homes, in their schools, and in their communities. It is, to some extent, a normalized aspect of social interactions. The high rates of violence are a testament to the country’s lingering legacy of inequality and exclusion and a symptom of its continuing fragility. While South Africa has made important gains in reducing violence in the post-Apartheid- era, it remains among the most violent countries in the world. After a significant decline during the period 1994–2011, annual homicide rates— the most commonly used proxy indicator for overall violence and crime—increased to 33 per 100,000 people, roughly 5 times the global average (ISS 2017b, UNODC 2013). Violence against women and children remains prevalent, and school violence presents a major challenge. Violence exacts enormous economic and social costs in South Africa. The direct costs of gender- based violence alone are an estimated R28.4– R48.2 billion (US$23–4 billion) per year (ISS 2017b).¹ A more recent study commissioned by Save the Children South Africa have estimated the direct and indirect costs of violence against children to be R238 billion (US$19 billion) per year (Save the Children 2016). Beyond the human impact of lives lost and altered by violence, the ¹ This estimate includes the direct costs of state agencies for policing, criminal justice, and the health sector only. It does not include indirect costs, such as lost earnings due to time not working or costs incurred by nonstate entities. 1 Contents Chapter 1 | Republic of South Africa Systematic Country Diagnostic costs of responding to violence overwhelm households and communities, drain state resources, and discourage investment in human and financial capital. To date, the policy response to crime and violence in South Africa has been uneven. However, opportunities for a more coherent response are emerging. Efforts are underway to streamline policies across departments and spheres of government and shift toward a better balance of measures to respond to violence, particularly through law enforcement, and to prevent it by addressing risk factors. South Africa possesses several elements that would facilitate such a process, including strong data collection systems, a growing evidence base for programs, and increasing dialogue among stakeholders. This paper is divided into seven sections. It begins with key definitions, and provides a typology of violence and crime, followed by an overview of crime and violence in South Africa. Next, it gives an overview of risk factors for the most prevalent types of crime and violence, with a focus on poverty and inequality, in line with the SCD framework. An overview of the key policies and institutions responsible for violence and crime prevention follows. The paper concludes with a few broad policy recommendations. 2 2 CRIME & VIOLENCE: KEY CONCEPTS This section is intended as an overview to some of the important concepts and definitions related to crime and violence. Because crime and violence cut across social, economic, political, and psychological dimensions, they are addressed by a variety of disciplines. This has given rise to a variety of definitions and concepts from the fields of criminology, psychology, economics, and even urban development. The terms violence and crime differ in important ways. The World Health Organization’s standard definition of violence is: “the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, which either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation” - (Krug et al 2002). A crime is an action that violates the law, regardless of whether or not it involves violence. White collar crimes are not violent, as an example. Furthermore, some acts of violence such as domestic and psychological violence, including stalking, are not considered crimes in many countries. Most development agencies, including the World Bank, have adopted the typology of violence put forward by the World Health Organization. It separates forms of violence into three categories based on the perpetrator’s status: 3 Contents Chapter 2 | Republic of South Africa Systematic Country Diagnostic 1 2 3 Self-inflicted violence, Interpersonal violence, Collective or political including attempted or perpetrated against a violence, committed to completed suicide, family member, community advance a particular self-abuse, or mutilation; member, or stranger by an social or political agenda, individual or small group, including organized such as a youth gang; crime, terrorism, war, and this includes domestic civil conflict. and intimate partner violence, most school- based violence, and most interpersonal common crimes committed by small groups or individuals; and Most World Bank violence-related operations have addressed interpersonal and collective violence. Work regarding self-inflicted violence has been limited to a few analytical pieces (see, for example, La Cava, Clert, and Lytle 2004). 7 4 3 VIOLENCE IN SOUTH AFRICA: DATA, DEFINITIONS, & KEY TRENDS South Africa has built relatively strong data collection systems on crime and violence, resulting in a solid understanding of incidence and prevalence, as well as research documenting different risk and protective factors. National government has implemented an annual Victims of Crime Survey (VOCS) since 1998 in order to inform government efforts on crime prevention and public safety. Data are representative at the provincial level and within provinces at the metro- non-metro levels. Since 2013, VOCS includes continuous data collection. The SAPS also collects data from official reports of crime, and publishes these annually, and a Demographic and Health Survey module for Violence Against Women was conducted in 2016. Each of the various sources have their own advantages and drawbacks in terms of reliability. In some cases, victimization surveys produce higher estimates than official sources— first, because many crimes are not reported to police; and second, because people are asked about incidents that may not fit the legal definition of a crime. Victimization surveys can also produce underestimates because people may not recall incidents or may be reluctant to disclose information to surveyors. Importantly, underreporting to authorities in South Africa can distort statistics, particularly around assault and sexual crimes (Stats SA 2016).² Reasons behind the underreporting vary. The Victims of Crime Survey 2014/15 asked victims why some did not report their assaults to police. The most common reasons for this were: 5 Contents Chapter 3 | Republic of South Africa Systematic Country Diagnostic “my family will resolve it,” “the police won’t do anything,” “I solved it myself,” and “it was not serious enough.” Victims of sexual offenses cited reasons such as it not being “serious enough,” “the police could do nothing,” and it being “inappropriate for police.” (Stats SA 2016). The most prevalent types of interpersonal violence in South Africa are homicide, violence against women, gender-based violence, intimate partner violence, violence against children, youth violence, and nonviolent crimes (see appendix A for definitions). Any typology is necessarily static. In reality, the different forms of violence overlap substantially. For example, sexual assault can be categorized as GBV, VAW or IPV depending on the circumstances. The discussion that follows focuses on the most prevalent and socially disruptive forms of crime and violence, as a means of understanding the general landscape of crime and violence in South Africa. More detailed analysis can be found in the annual VOCS, the Demographic and Health Surveys (DHS) (for VAW) and the National Youth Lifestyle Study (for violence by and against youth). 3.1. HOMICIDE Homicide is often used as a proxy indicator² for overall levels of crime and violence. While it is, in many ways, an imperfect indicator, homicides are usually the most visible manifestation of crime, and data tend to be more reliable than for other crimes. Where homicide rates are high, other forms of violent crime, tend to be high as well. At the national level in South Africa, homicides remain high, and are on the rise since 2012. After showing important declines since the end of Apartheid, homicides rates have increased each year since 2012. The 2013 Global Study on Homicide (UNODC 2013) showed that South Africa’s homicide rates decreased between 1995 and 2011 by more than 50% from 64.9 to 30.0 per 100,000 habitants. Similarly, most categories of reported crimes showed decreased tendencies between 2004 and 2011 according to SAPS’ statistics. However, since hitting a low of 30.0 in 2011, homicide rates have increased. Between 2015/16 and 2016/17, the homicide rate rose marginally from 34 to 34.1 per 100,000 people, with an average of 52.1 people murdered per day (Graph 1, SAPS 2017). Figure 3.1. Homicide Rates in South Africa, 2007/8–2016/17 40,000 37.6 36.4 40.0 33.3 31.9 32.9 34.0 34.1 35,000 31.1 30.1 30.9 35.0 Actual : Murder 30,000 30.0 Ratio : Murder 1.8% 25,000 25.0 20,000 20.0 15,000 18,400 18,084 16,767 15,893 15,554 16,213 17,023 17,805 18,673 19,016 15.0 10,000 10.0 5,000 5.0 0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 0.0 Ratio Murder Source: SAPS 2017. ² As a proxy, homicide data have important limitations. Different sectors, such as criminal justice and health, might define homicide differently or might not report homicides at all. For example, if a person dies in the hospital after an assault, the hospital might not report the death as homicide. Definitions also vary across countries; some define abortion as homicide, for example. Finally, because homicide is an extreme and rare event, it cannot provide a full picture of the types of crime and their intensity in a given society. 6 Republic of South Africa Systematic Country Diagnostic | Chapter 3 As figure 3.2 illustrates, homicide rates vary by province. The lowest rates are found in Limpopo and Mpumalanga and the highest in Eastern and Western Cape. The provinces with the greatest increases in homicide rates between 2016 and 2017 were Mpumlanga at 11.1 percent and Gauteng at 6.7 percent; Limpopo and Northern Cape experienced a decrease of 9.5 and 7.5 percent, respectively (SAPS 2017). Figure 3.2. Homicide Rates in South Africa by Province, 2017 Limpopo Mpumalanga North West Northern Cape Gauteng Free State National KwaZulu-Natal Western Cape Eastern Cape 0 10 20 30 40 50 60 Source: SAPS 2017. 3.2. VIOLENCE AGAINST WOMEN The physical and sexual abuse of women is among the most prevalent manifestations of violence in South Africa. Underreporting is a major concern; many researchers have posited that actual rates may be much higher than official assessments (Jewkes et al. 2011, 2015; Machisa et al. 2011, Matthews et al. 2016). Victimization surveys using specialized methodologies allow for more reliable estimates.³ The DHS asks women if they have experienced physical or sexual violence in their lifetimes and in the previous 12 months. The 2016 survey reported that 21 percent of ever-partnered women had experienced abuse by an intimate partner at some point; and 8 percent reported this type of violence in the previous 12 months. Six percent reported having experienced sexual violence at the hand of an intimate partner in their lifetime, and 2 percent reported such violence over the previous 12 months (Stats SA 2017). These rates are roughly consistent with global averages. The World Health Organization has estimated that 30 percent of women worldwide have been the victim of physical or sexual violence by an intimate partner (WHO 2013). ³ The South African Police Service does not disaggregate data on rape or domestic violence from that on assaults. Further, these types of crimes are historically underreported in South Africa. Most estimates of violence against women and children are therefore taken from victimization surveys, which use a specialized methodology to increase reliability. 7 Contents Chapter 3 | Republic of South Africa Systematic Country Diagnostic Women’s experience of physical violence and sexual violence shows variation across geographic regions. Reported physical violence (ever in one’s lifetime) ranged from a low of 14 percent in KwaZulu-Natal to a high of 32 percent in the Eastern Cape. No significant difference was observed between women in urban (20.2 percent) and rural (20.9) areas overall. However, there were variations in experience of sexual violence by geographic area, ranging from a low of 3.1 percent in KwaZulu- Natal to a high of 11.8 in North West (Stats SA 2017, p. 57). The percentage of women reporting sexual violence by an intimate partner showed less variation by employment status (8 percent of employed women versus 5.1 percent unemployed) or by education level (4.7 percent for women with no formal education versus 4.7 for women with more than secondary level education). This is discussed in more detail below. Surveys that ask men to self-report acts of sexual and physical violence in South Africa suggest more alarming figures. Multiple Perpetrator Rape (MPR), involving one victim coerced by two or more perpetrators, appears to be highly prevalent in South Africa. MPR is a learned activity, and usually perpetrated as part of established cultural practices rooted in imposing gender discipline (Jewkes et al 2011). In a 2008 survey (N=1686) of men aged 18 and older across three districts, 27.6 percent reported ever having raped a woman, with 8.8 percent of these reporting they engaged in MPR (Jewkes et al 2015). Multiple perpetrator rape, which involves one victim coerced by two or more perpetrators, appears to be highly prevalent in South Africa. 8 Republic of South Africa Systematic Country Diagnostic | Chapter 3 3.3 VIOLENCE AGAINST CHILDREN Violence against children under 18 years of age is highly prevalent in South Africa. Underreporting is a serious problem, such that actual levels are suspected to be much higher than official numbers. A study using 2009 data from medicolegal laboratories estimated that South Africa’s homicide rate among children (under age 18) was 5.5 per 100,000, more than double the World Health Organization’s global average of 2.4. Rates for boys were much higher than for girls, at 6.9 and 3.9, respectively. Abuse and neglect preceded 44.5 percent of homicides, and girls were three times more likely than boys to suffer abuse and neglect prior to their death (Matthews et al. 2013). One nationally representative survey on child abuse and neglect found that one in five South African children had experienced sexual abuse in his or her lifetime, and one in three had been physically abused (Save the Children 2016). Almost one third of South African children had experienced violence in their homes (UBS Optimus 2015). Children who experience these forms of violence are more inclined than others to suffer from depression and anxiety—or to turn to violence themselves (Matthews et al. 2016). They are also likely to earn less than their peers when they enter the job market, and are more likely to use drugs or abuse alcohol (UBS Optimus 2015; Fang et al 2016). (UBS Optimus 2015; Fang et al. 2016). 3.4. YOUTH VIOLENCE Youth are more vulnerable to violence than adults. The National Youth Lifestyle Study, conducted by the Centre for Justice and Crime Prevention, reported that youth (ages 12–22) are at significant risk for being victimized by a crime (CJCP 2008). According to the comparable, adult-focused National Crime and Victimization Survey (ISS 2003), more than one-quarter of South African youth—26 percent—had experienced violence in the preceding 12 months—five times the 5 percent rate of adults. Reported rates for assault, robbery, and sexual violence are all higher than those of adults (CJCP 2008). More recently, a 2016 study estimated that 35 percent of youth ages 15 to 17 have experienced sexual and/or physical abuse (USAHRC and UNICEF 2016). Rates of lethal violence were much higher among young boys than girls. Homicide rates in 2009 among boys ages 15–17 were estimated at 21.7 per 100,000 compared with 4.6 among girls of the same age cohort (Matthews et al. 2013). A significant share of violent encounters among youth occur at school, especially in the classroom. According to the 2012 National School Violence Survey, 22.2 percent of secondary school children have experienced threats of violence, assault, robbery, or sexual assault—including rape—while at school. The classroom is the most frequent site of violent incidents, and female learners are more likely to be victimized than their male counterparts (Burton and Leoschut 2013). Roughly 90 percent of threats, sexual assaults, and thefts, as well as 69.8 percent of assaults, are committed by other students while attending school (Burton and Leoschut 2013). Violence against children and youth generates enormous costs, not only in the immediate term, but also over the life cycle. Among the most troubling challenges is the increased propensity among those who have experienced violence in childhood to then perpetrate violence later in life. The accrued cost is staggering—psychologically, emotionally, and monetarily, the latter through both direct and 9 Contents Chapter 3 | Republic of South Africa Systematic Country Diagnostic indirect costs (e.g., lost earnings). A 2015 study estimates the loss of human capital for people who have experienced violence in childhood at R238 billion, (USD18bn) an amount equivalent to 6 percent of gross domestic product and more than double the state’s current annual spending on the criminal justice system (Save the Children 2016). The most common crimes against households were home break-in or burglary at 53% 3.5. NON-VIOLENT CRIMES The Victims of Crime Survey (VOCS) is the primary source for data on non-violent crime. The VOCS collects data on household crimes, and crimes against individuals. Household crimes include: motor vehicle, livestock, crop, and bicycle theft; theft out of a motor vehicle; burglary; robbery; murder; trafficking in persons; destruction of buildings; and vandalism. Crimes against individuals include: theft of personal property, hijacking of a motor vehicle, robbery, sexual offense,⁴ assault, consumer fraud, and corruption (Stats SA 2018). According to VOCS, the overall crime experienced by South Africans ages 16 and older declined slightly in 2016–17 from previous years (Table 3.1). The most common crimes against households were home break-in or burglary (53 percent), followed by theft of livestock (11 percent) and robbery (10 percent). Male-headed households were slightly more likely to be victimized (7.5 percent) than female-headed households (6.6 percent). Most individual crimes declined in 2016, with important exceptions, including sexual offenses and the hijacking of motor vehicles. The rate of individuals experiencing repeated victimization has increased, which suggests that crime may be heavily concentrated in fewer households (Stats SA 2018). ⁴ Sexual offences include rape, grabbing or touching without consent. 10 Republic of South Africa Systematic Country Diagnostic | Chapter 3 Table 3.1. Five-year trends in percentage of households and individuals reporting crime victimization 2013-2017 Household crime 2013/14 2014/15 2015/16 2016/17 Qty Western Cape 13,4 13,9 11,1 9,0 Eastern Cape 12,3 12,4 10,2 9,0 Northern Cape 10,7 10,0 7,5 8,2 Free State 6,8 6,9 7,2 5,7 KwaZulu-Natal 10,5 9,1 7,5 7,8 North West 9,9 7,2 7,2 6,3 Gauteng 9,7 10,2 9,1 6,5 Mpumalanga 11,5 9,7 9,0 6,5 Limpopo 6,5 6,6 5,0 5,4 South Africa 10,2 9,9 8,5 7,2 MVI 109 107 109 106 Individual crime Western Cape 7,6 7,0 6,5 5,4 Eastern Cape 4,8 4,3 5,2 5,1 Northern Cape 5,3 5,9 4,3 2,9 Free State 5,0 4,8 3,3 3,5 KwaZulu-Natal 3,2 3,1 2,3 2,3 North West 4,9 4,5 3,6 3,1 Gauteng 4,5 5,0 3,9 3,1 Mpumalanga 5,2 3,5 4,3 4,1 Limpopo 3,0 2,0 2,0 2,3 South Africa 4,7 4,4 3,9 3,5 MVI 104 105 105 105 11 Contents Chapter 3 | Republic of South Africa Systematic Country Diagnostic 3.6. CRIMES AGAINST BUSINESSES Crimes against businesses exact an enormous cost, especially on smaller firms. The Consumer Goods Crime Risk Initiative⁵ reported that between 2013 and 2014, armed robberies at shopping malls increased by a staggering 142 percent, from 274 to 665, respectively. The cost of crime to firms in 2016 represented about 20 percent of their total annual revenues; this is much higher than many other countries in the upper-middle income group (ISS 2016). Direct losses, including stolen goods and inventory due to theft and carjacking, as well as direct security-related costs, translate into reduced profits and less investment in business infrastructure. Indirect costs include lost tourism, especially in townships with higher crime rates (Ushakov, 2014). The high cost of security is a serious barrier for smaller firms. The National Treasury of South Africa (2016) (Mahofa et al 2016) estimated that a 1 percent increase in total crime reduced business entry by 0.53 percent. Similarly, a study by Bhorat and colleagues (2017) showed that an increase in crimes against businesses in a municipal unit was associated with a reduced growth in the stock of active firms, primarily due to their having exited. A study by Edwards and Sundaram (2013) recorded similar observations. They found that a 10 percent increase in the number of severe crimes was associated with a 1.5 percent reduction in the number of start-up firms. Small and medium enterprises operating in townships consider crime to be the most critical constraint to investment and growth (World Bank 2015). The 2012 World Bank Diepsloot Enterprise Survey data confirms this: almost 27 percent of surveyed firms in the urban township of Diepsloot ranked crime as the most serious obstacle to businesses. In 2012, 90 firms—20 percent of the all firms in the township—had experienced negative crime-related shocks. Small and medium enterprises had been disproportionally affected. 3.7. PERCEPTIONS OF INSECURITY AND THEIR IMPACT ON BEHAVIOR According to the Victims of Crime Survey, the overall decline in crime has not been accompanied by improved perceptions of safety. The percentage of households victimized by crime declined from 9 percent in 2015/16 to 7 percent in 2016/17, but the percentage of people feeling safe while walking alone in their neighborhood decreased from 31 to 29 percent over the same time period. Perceptions of safety change with the time of day: about 84.8 percent of households feel safe during daylight, but only 29.4 percent do after dark. An in-depth analysis of the survey for the period 2011–14 reveals that, even though the rate of crime had been declining, South Africans perceived it to be increasing (Stats SA 2016). The rate of South African households that experienced a home burglary or robbery between 2010 and 2016 declined from 6.8 to 5.7 percent, but the perception that crime had been increasing rose from 31.2 percent in 2010 to 41.7 percent in 2015/16 (Stats SA 2016). According to the same survey, people change their behaviors to avoid being victimized by crime (table 3.2). One third (31.5 percent) of the surveyed households refrain from occupying open spaces, and nearly one fifth (19.9 percent) do not allow their children to play outside because they fear crime in their neighborhood. ⁵ The initiative was established in 2012 as an industry-wide collaborative mechanism focused on preventing and mitigating the significant crime risk against the consumer goods and retail industry. 12 Republic of South Africa Systematic Country Diagnostic | Chapter 3 Table 3.2. Percentage of households who are prevented from engaging in daily activities due to crime in their neighborhood, 2016/17⁶ Households prevented from engaging in daily activities Activity Number in thousands Percent Going to open spaces 4,820 31,5 Allowing children to play in area 2,458 19,9 Walk to town 2,026 15,0 Dressing in any way 1,802 11,3 Walk to shops 1,770 11,3 Public transport 1,747 11,5 Allowing children to walk to school 1,668 13,9 Starting a home business 1,623 12,1 Expressing sexual orientation 1,489 9,3 Keeping livestock 846 11,1 Walking to fetch water 549 6,0 Source: Stats SA 2018: 58. ⁶ Source: VOCS Table 53, p. 58. (Stats SA 2018) 13 4 SPATIAL DISTRIBUTION OF CRIME & VIOLENCE Crime in South Africa is concentrated in specific geographic areas. A long-term trend analysis is difficult due to the absence of spatial data for earlier years, but in 2016, the government of South Africa merged spatial data from the VOCS with that of the SAPS, enabling an analysis by police station boundary. These data were then merged with 2011 census data to identify locations where crimes are frequently committed as well as hot spots within particular communities. The following in-depth geospatial analysis draws on data regarding contact crimes: assault, sexual offense, and murder (Stats SA 2016). The locations of assaults tend to differ among male and female victims, partly due to differences in mobility and frequented sites (see figure 4.1). Women are much more likely to be assaulted in the home than men; while men are more likely to be assaulted in the street or at a place of entertainment, such as a bar or restaurant. These differences may also be influenced by the type of assault: women are at a higher risk of being victimized by an intimate partner in the home, while males are more likely to be exposed to violence in a public setting. Women are much more likely to be assaulted in the home than men. 14 Republic of South Africa Systematic Country Diagnostic | Chapter 4 Figure 4.1. Time‐series analysis of places where assault occurred by sex (per cent), VOCS 2011‐2014/15 100,0 90,0 80,0 70,0 Percentage 60,0 50,0 40,0 30,0 20,0 10,0 0,0 Male Female Male Female Male Female 2011 2013/14 2014/15 At home 20,1 56.5 9,2 44,0 23,6 57,8 In the street 38,4 23,6 67,5 33,3 45,1 25,8 Entertainment place 20,1 5,9 12,6 12, 13,7 2,6 Workplace 7,0 2,0 5,3 0,0 2,2 7,8 Outdoors 5,4 1,6 0,0 5,4 4,3 4,1 Other 9,0 10,4 5,4 5,1 11,1 1,9 Source: Stats SA 2016: 13. Assault and murder rates are higher in urban areas. The highest prevalence of assault is in KwalZulu- Natal, Western Cape, and Eastern Cape. Assault rates are highest within metropolitan areas, including Cape Town, Nelson Mandela Bay, and eThekwini. Even in non-metropolitan municipalities, hot spots of crime are concentrated in the more urbanized areas, such as Mangaung, Polokwane, and Sol Plaatjie. The murder hot spots are in Cape Town, eThekwini, Mnquma, Sol Plaaitjie, Nala City of Matlosana, Polokwane, Emalahleni, and Westonaria (Stats SA 2016). 15 5 RISK FACTORS FOR VICTIMIZATION & PERPETRATION OF VIOLENCE AND CRIME IN SOUTH AFRICA As elsewhere, South African’s current violence cannot be explained by a single factor or experience. There is broad consensus among policy experts and researchers of crime and violence prevention that violent behavior is provoked by a complex interaction of physiological, psychological, and environmental factors. There are, therefore, no direct “causes” of violence; there are instead individual characteristics related to biology, personality, and environment that increase a person’s risk of perpetrating or experiencing violence. An accumulation of these risk factors is associated with an increased tendency toward being a victim or a perpetrator of violence. On the other hand, protective factors characterize resilient individuals and environments in the presence of violence. Figure 5.1. Ecological Model SOCIETAL COMMUNITY RELATIONSHIP INDIVIDUAL 16 Republic of South Africa Systematic Country Diagnostic | Chapter 5 The most commonly used model to understand risk factors of violence is the ecological framework (figure 5.1),⁷ which conceptualizes violence as the outcome of a variety of interrelated factors at the individual, interpersonal, community, and societal levels. The more risk factors that are present, and the more levels involved, the greater the propensity for violence. Risk and protective factors can change over the course of a lifetime. Children, for example, are more impacted by individual and relationship factors, but youth may be more influenced by community-level risk factors (Chioda 2017). Further, while risk factors for different types of violence often overlap in the life experiences of both victims and perpetrators, certain risk factors may be more salient to particular types of crime. The combination of risk and protective factors is unique to each individual, which helps explain why some people and communities survive—and even thrive—in high-risk environments, while others do not. There is no mechanical relationship between these factors and outcomes: risk factors do not cause violence, and protective factors do not prevent it. Instead, they influence the way individuals and communities respond to it. The literature on risk factors for various types of violence in South Africa is particularly rich. A 2016 study prepared for the Inter-Ministerial Committee on Violence Against Women and Children (Matthews et al. 2016) included 20 datasets that identify a set of common risk factors for victims and perpetrators of violence. While the committee is specifically focused on violence against women and children, the identified risk factors are consistent with those associated with other forms of violence. A comprehensive discussion of each factor is beyond the scope of this paper, but table 5.1 provides a summary of factors, and a more detailed discussion of some of the more relevant ones follows. Table 5.1. Key Risk and Protective Factors for Violence in South Africa Level Definition Risk Factors Protective Factors Individual Biological, psychological, Witnessing violence in Socioemotional skills to and behavioral the home or community; communicate and cope characteristics beliefs that condone with stress; beliefs that interpersonal violence; promote nonviolent poverty; age; gender; conflict resolution alcohol or drug use Relationship Influence of family Patriarchal or Safe, stable, and or caregiver (family physically violent family nonviolent relationships environment and environment; substance in the home; healthy parental behavior); use by parents household-level relationships with peers communication and/or intimate partners ⁷ For an overview, see www.who.int/violenceprevention/approach/ecology/en. 17 Chapter 5 | Republic of South Africa Systematic Country Diagnostic Level Definition Risk Factors Protective Factors Community Relationships Poor policing; Access to secure spaces; with schools and access to firearms, community norms neighborhood; social drugs, and alcohol; that sanction violent integration and capital poverty; tolerance of behavior; healthy and interpersonal violence positive recreational opportunities (e.g., sports and social groups) Society Demographic and Inadequate laws Social norms that social changes, income governing firearms; promote nonviolent inequality, and political policies that maintain conflict resolution; structures; cultural inequality or policies to improve influences and norms; discrimination based social, economic, and economic and social on race or gender; political inclusion policies culture of male sexual entitlement; pro-violence norms Sources: Matthews et al. 2016, WHO 2002, Jewkes et al. 2015, Machisa et al. 2011. 5.1. INDIVIDUAL LEVEL At the individual level, directly experiencing or witnessing violence in the home is a central risk factor for becoming a victim or a perpetrator later in life. Girls are particularly at risk of being victimized in an intimate relationship. Boys are more likely to behave violently at school or in the community, and as adults they are more likely to be violent with an intimate partner. The level of risk of being victimized by or perpetrating violence varies over a lifetime, but it peaks during youth. Gender also impacts vulnerability: men and boys are more likely to perpetrate violence or to be injured or killed while participating in a violent act, while women and girls are more likely to be victimized by violence at the hands of a family member or intimate partner (WHO 2013). As a review by the Department of Social Development (DSD, SASSA, and UNICEF 2012) found, while acts of violence can occur in any setting, they tend to be concentrated in poorer, urban areas. Poverty increases vulnerability to violence and makes it more difficult for the victim and the household to deal with its impacts. Because poverty is a significant risk factor, poverty reduction measures have the potential to reduce vulnerability to violence. The primary tool for economic assistance in South Africa is the Child Support Grant or the Care Dependency Grant for children with disabilities. Existing evidence suggests that these programs have improved the well-being of children in terms of nutrition, health, and education outcomes and access to early childhood development programs; however, they have not yet been evaluated in terms of their impact on exposure to violence (DSD, SASSA, and UNICEF 2012). Because 18 Republic of South Africa Systematic Country Diagnostic | Chapter 5 the grants impact other protective factors, they might be able to support positive outcomes regarding violence as well. The relationship between employment status and violence is complex. In qualitative studies, people, especially youth, have consistently identified the lack of access to steady employment, income, and educational opportunities as drivers of violence (Ward 2006, World Bank 2010). However, strong empirical work that links these elements is lacking; and some employment programs have aggravated conflict. The Community Work Program implemented in Cape Town resulted in an increase of violence against women by men; Bruce (2015) proposed the possible explanation that women had benefitted more than men from the job creation. Regarding violence against women in particular, employment and education show no clear relationship. The Stats SA/DHS survey (Stats SA 2017) reports only slight differences in women’s experience of physical violence by employment status, with 22.4 percent of employed women and 19.2 percent of unemployed women reporting ever having experienced physical violence by a partner. Experience of physical abuse does seem to vary with education levels; 21 percent of women with no education reported abuse in their lifetime, compared to 17.3 percent who had secondary education and 12.4 percent with more than secondary level education (Stats SA 2017, p.55). These numbers are roughly consistent with international experience, which show no clear relationship between women’s employment status and risk of violence by an intimate partner (WHO 2013). In some cases, employment may provide an exit opportunity from an unhealthy relationship, while in others, the income and/or time spent away from home while working may contribute to conflict in the relationship. Similarly, some studies have shown a relationship between education levels and risk of violence, because lower The income and/ or time spent away from home while working may contribute to conflict in relationships. 19 Chapter 5 | Republic of South Africa Systematic Country Diagnostic levels of formal education correlate with greater economic dependency. By this token, women who have more access to formal education may be better able to challenge patriarchal norms or leave a violent relationship. This is consistent with the global literature on intimate partner violence, which has shown no clear relationship between employment, or education, and experience of violence. Overall, access to a job or income can increase conflict in the household (bringing more violence) or it can support women to leave an abusive situation (reducing violence). The key seems to be how much autonomy women have over the resources (WHO 2013, Willman 2009). Unemployment and education levels may increase risk of perpetration of resource acquisition crimes such as property crime and petty burglary. In a study using 2011 census data and precinct- level South African Police Service data, Bhorat and colleagues (2017) explored the relationships between income, intra-precinct inequality, and unemployment with three categories of crime: property crimes; burglary; and violent crimes, including murder, sexual assault, common assault, and grievous bodily harm. The study first found that unemployment was not related to any of categories. Property crime seemed to be most sensitive to economic factors because high levels of inequality signaled higher potential returns; it increased up to a certain level of income and then leveled off when the risk of arrest outweighed the perceived economic benefits derived from committing the crime. There was not a strong relationship between robbery and any of the economic factors. The study’s authors concluded that economic factors were quite limited in explaining violent crime in South Africa, noting that violent crime was driven by motivations “more complex and psychological than those behind resource acquisition crime.” (Bhorat et al. 2016: 35).⁸ Some individual-level risk factors appear more salient than others in predicting perpetrators of sexual violence in South Africa. In one study (N=1,686), men who reported having engaged in sexual violence shared significantly different experiences than those of men who did not (Jewkes et al. 2015). The researchers found no age-related differences, but did find that men with higher earnings were more likely to perpetrate multiple perpetrator rape; and men who reported having raped earned over R500 per month more than those who did not. In addition, the men who reported having committed rape tended to have been raised by mothers with higher levels of education and their fathers tended to be absent throughout their childhood. These same men were more likely to report that they had been forced into sex by a man during their own childhood: 16.2 percent of men who reported having raped a woman and 19.5 percent of men who reported having participated in multiple perpetrator rape made this claim compared with 6.3 percent of the men who asserted that they had never committed rape against a woman. These findings underscore the need for violence prevention over the course of a lifetime (Chioda 2017). 5.2. RELATIONSHIP LEVEL At the family level, positive and secure relationships between a child and caretakers are essential to social and physical development, constituting key protective factors against violence. Such relationships provide children with safe learning environments where they can develop the cognitive, ⁸ The study findings may have been influenced by the reliance on precinct-level crime statistics as reporting rates tend to vary among wealthier and poorer areas. 20 Republic of South Africa Systematic Country Diagnostic | Chapter 5 social, and emotional skills they need to interact in social contexts and to establish positive social relationships (Mathers et al. 2012). A child living in a home without healthy family relationships or a safe and nurturing environment is significantly more likely to be maltreated than one in home that is positive and stable (McCauley 1998). The high incidence of single-headed households and family disruption in poorer urban areas are legacies of the exclusive policies that separated families during apartheid; and it is associated with increased risk of being victimized by and perpetrating violence (Coovadia et al. 2009). Other important risk factors for being victimized include substance abuse in the home, a disruptive family structure, incarceration of a family member, and a family culture of harsh discipline or one where conflict is often resolved with violence. 5.3. COMMUNITY LEVEL At the community level, environmental factors such as the availability of weapons, drugs, and alcohol impact propensity for violence (Machisa et al. 2011, Matthews et al. 2016). According to the 2014/15 Victims of Crime Survey, the most common weapon used during an assault in South Africa was a knife, although the use of firearms appeared to be increasing (Stats SA 2016). Access to alcohol appears to be a key risk factor for various types of violence. One review reported that children who had been exposed to adults abusing alcohol are more at risk of being victimized by or perpetrating violence if they also drink alcohol. Boys are more at risk than girls on both counts (Matthews et al. 2016). These findings are underscored by global evidence that measures to limit alcohol consumption help decrease violence levels. Such measures have been dramatically successful in the cities of other middle-income countries, such as Diadema in Brazil⁹ and Cali in Colombia (Sánchez et al. 2011). Taxes on alcohol have also reaped positive impacts.¹⁰ In South Africa, a number of risk and protective factors combine at the community level. A qualitative study on gangs and community violence in various Cape Town communities with high levels of gang activity asked residents to identify the key risk and protective factors for young people. In the focus groups, children identified poverty, poor neighborhood policing, high levels of drug activity and violence, a lack of alternative recreational activities, and poor television role models as risk factors for gang involvement. In addition, maternal economic dependence, overcrowded homes, and parental stress and frustration have placed children and adolescents at increased risk of involvement in violence or gangs (Ward 2006). Important protective factors emerged, including: (1) poverty alleviation through employment, (2) community cohesion, (3) lower levels of neighborhood crime and drug activity, (4) higher levels of school enrollment, (5) more religious institutions, and (6) access to recreational facilities and extramural activities (Ward 2006). No strong empirical studies have yet explored whether or not these factors do indeed protect against gang or other forms of community violence; additional research is needed. ⁹ Restrictions on alcohol sales during critical hours were associated with a 44 percent reduction in homicides (Duailibi et al. 2007). ¹⁰ A 2010 meta-analysis systematically reviewed 50 articles containing 340 estimates for the relationship between alcohol taxes and prices to measures of risky behavior and mortality. Findings were consistent with the wealth of older studies, suggesting that tax increases on alcoholic beverages resulted in an overall estimated decrease in violence of over 2 percent and a decrease in crime by 1.4 percent (Wagenaar 2009). 21 Chapter 5 | Republic of South Africa Systematic Country Diagnostic Access to safe public spaces is a key protective factor at the community level. Violence Prevention through Urban Upgrading—a joint initiative of the government of Cape Town, the German Development Bank, the Khayelitsha community, and various community and nongovernmental partners—supports townships that are experiencing high levels of violence. The initiative has focused on the redesign of township areas to promote safety and offer safe spaces for community activities (Western Cape Government 2014). It has also sought to reduce the social exclusion created under the apartheid regime (Cassidy et al. 2015). The initiative has created safe node area committees in each target community and has encouraged community leadership in project activities. The project is currently under evaluation (Cassidy et al. 2015). Evidence suggests that supporting access to safe schools and encouraging involvement in afterschool activities serve as protective factors for students in South Africa, in some cases compensating for the effects of their exposure to violence. Ward and colleagues, using structural equation modeling to explore the impact of afterschool activities and school supports on mental health and behavior, found that children who had received support at school and who had participated in afterschool activities were reported to have less anxiety, depression, and behavioral problems than the control group (Ward et al. 2007). Similarly, a study by Cluver and colleagues found that students who had attended afterschool sports, music, or dance groups experienced less bullying than the comparison group (Cluver, Bowes, and Gardner: 10). These activities had their greatest impact when young people were involved in decisions about what types of programs were provided as well as where and when they would be held (Matthews et al. 2016). 5.4. SOCIETAL LEVEL At the societal level, South Africa is home to some of the highest inequality in the world. Vertical inequalities – overall disparities in income – are stark. The poorest 20% of the population consume less than 3% of total expenditures, while the wealthiest 20% consume 65% (Statistics South Africa 2016). While vertical inequalities merit attention for a variety of reasons, research from a wide variety of contexts shows that horizontal inequalities (HIs) – those that accumulate across social groups, including along racial or ethnic lines – are much more likely to be associated with social instability, and in some cases, violence (Stewart 2002, Stewart and Langer 2007). Exclusionary systems that are perceived to privilege some groups at the expense of others,¹¹ create fertile ground for violence. The empirical relationship is especially strong when HIs accumulate across different realms (economic and political, for example) (Justino, 2017; Østby 2008)¹² It is important to note that objective HIs themselves do not automatically or inevitably translate to violence, or even generate strong grievances in many cases. Many social groups may feel excluded or may objectively suffer from exclusion; inequality is present in most, if not all, countries. Instead, horizontal inequalities increase the risk of violence when strong perceptions of injustice attach to HIs, especially when they allow for assigning blame for the inequality to a rival group (PRIO 2017). These ¹¹ Galtung, 1969, defines these conditions as “structural violence” as discussed above. ¹² On education, see UNICEF 2015; on infant mortality (for Indonesia only) see Ostby, Urdal, Tadjoeddin, Murshe, Strand 2011; Cederman et al, 2013. 22 Republic of South Africa Systematic Country Diagnostic | Chapter 5 perceptions can be mobilized by individuals, and commonly by social leaders – from gang leaders to armed groups – to incite and justify violence (UN and World Bank 2018).¹³ The longer the HIs persist, the harder they are to reverse, often requiring direct government intervention to remedy them. Some examples of this include affirmative action to redistribute resources through economic development programs, land reform, or quotas for historically disadvantaged groups in public universities. In many cases, actions taken to remedy HIs can themselves provoke a backlash by those who perceive themselves to be losing out (Mitra and Ray 2014). One often cited example is Sri Lanka, where policies to reduce horizontal inequalities between the Tamil minority and the disadvantaged Sinhalese majority during the 1960s and 1970s contributed to a sense of disempowerment for the Tamil group. Studies of affirmative action in African countries have shown that the more successful programs were accompanied by measures that increase political support for the reforms, and tolerance for the groups that benefit (Langer, Stewart and Schroyens, 2016). As inequality among black South Africans increases, the issue of HIs becomes more important as a source of instability and a risk factor for violence. As detailed in the SCD, while social progress has been significant since 1994, poverty has persisted among particular groups, and much poverty is chronic: 60 percent of South Africa’s poor were classified as poor in all four household surveys undertaken between 2008 and 2015, while nearly 80 percent were poor at least once in that period, according to the 2018 World Bank Poverty Assessment (World Bank SCD, p18). By itself, these trends may not inevitably produce violence, but if they are mobilized as grievances, the risk increases. Research has documented how past social exclusion blocks current pathways to social mobility by limiting the access of certain groups to social networks outside their own group. Adato, Carter, and May (2006), using panel data for the period 1993–98, finds that while poor households possess ample social capital (also called bonding social capital) within their smaller identity groups, they have limited access to bridging social capital with other groups, that could enable them to acquire assets and achieve upward mobility. The failure of democracy to fulfil many of the expectations for greater inclusion and opportunity is a source of frustration and grievance. The aspirations created by increasing education in the Homelands during the 1980s and 1990s were simply not met by broader opportunities for socio- economic advancement. For many, this constrained the transition from youth to adult and underlie the problems with crime and violence in South Africa today (Beinart, Delius, and Hay, 2017). Measures to increase inclusion particularly through addressing horizontal inequalities, act as an important protective factor against violence and instability at the societal level. Hegre et al. (2016) examined data from all countries between 1960 and 2013,¹⁴ finding that countries with higher levels of inequality face greater challenges in mitigating the risk of violence. Similarly, Min et al. (2017) find that countries with policies to increase the participation of previously excluded groups, to influence government policy, and to increase political engagement during economic downturns experience less violence. Gender inclusion, in particular, shows a robust, empirical relationship with lower levels of ¹³ See Chapter 4 of Pathways for Peace report (UN and World Bank 2018) for a comprehensive discussion of the relationships between inequality, grievances, and mobilization to violence. ¹⁴ The study drew on the Upsaala Conflict Data Program (UCDP) data set and on a set of five scenarios for policy choices from the Shared Socio-Economic Pathways initiative. See O’Neill et al. 2014 for a description of the study and dataset. 23 Chapter 5 | Republic of South Africa Systematic Country Diagnostic violence, from the local to the international level (Herbert 2017; Hudson et al. 2009, Caprioli et al. 2007, Caprioli and Tumbore 2003). Governments of countries with more equitable gender relations, as measured by levels of violence against women, labor market participation and income disparities, for example, are significantly less likely to initiate interstate conflict or escalate civil conflict (Hudson 2012). In contrast, countries with higher levels of gender inequality are associated not only with increased risk of international or civil war, but also with more severe levels of violence used in conflict (Caprioli and Boyer 2001). As South Africa continues to grow, there is a risk that inequality will continue to increase; thus measures to promote inclusion could help protect against instability and the risk of violence. Social norms that sanction against violence are a key protective factor at the societal level, and the government has recognized them as a critical area for change. In its 2012 review, the Department of Social Development noted that the prevalence of violence in South African society today can largely be explained as a legacy of the apartheid system. During apartheid, violence was sanctioned from the top levels of society. The report asserted that those in power today have a responsibility to reverse this trend by promoting healthier social norms. South Africa has important experience in strengthening social norms against violence. South African organizations have generated models for normative change that have served as examples for effective prevention around the world, such as the Soul City ‘edutainment’ project, a soap opera that sought to change attitudes about intimate partner violence. An evaluation of the program’s fourth season using a random, national sample, found a significant decrease in those who held favourable attitudes about violence toward women (Usdin et al 2005).¹⁵ Other programs to promote healthier social norms in South Africa have combined economic empowerment with activities to strengthen socio-emotional skills as protective factors. A South African microfinance program with women from the poorest household in rural areas combined financing with skill-building sessions on HIV and GBV prevention and encouraged participation of men and boys. A randomized control trial found 55% lower rates of intimate partner violence in intervention communities than in the control group. It also reports enhanced household communication and attitude that challenge gender roles.¹⁶ These provide an important base that could be scaled up. ¹⁵ Notably, the evaluation did not include questions in changes in behavior, but only assessed changes in attitudes. The extent to which shifting attitudes contribute to actual behavioral change is unclear, though there is some evidence that increasing social sanctions against violence help act as a deterrent. ¹⁶ Pronyk PM et al. “Effect of a structural intervention for the prevention of intimate-partner violence and HIV in rural South Africa: a cluster randomised trial.” Lancet 368 (2006):1973–83. 24 6 THE STATE’S RESPONSE: AN OVERVIEW The policy landscape for crime and violence prevention in South Africa is tremendously complex. Responsibility for prevention is spread across 11 government departments, and nongovernmental organizations (NGOs) deliver many services. Several departments have conducted policy making around crime and violence prevention largely in isolation from one another, which has resulted in a lack of coherence and limited impact. To fully understand the policy landscape and how it is changing, a more comprehensive assessment is needed. Because some key policies are currently under review, relevant documents were unavailable at the time of this writing. This following overview is based on available records as well as interviews with local experts.¹⁷ 6.1. KEY LEGISLATION The overarching policy governing the state response to crime and violence is laid out in the National Development Plan 2012- 2030. Drawing on the South African constitution’s bill of rights, which protects the rights of freedom and security of persons, the plan defines the responsibilities of various government departments regarding community safety. It specifies 14 outcomes, which reflect developmental impacts to meet ¹⁷ In addition to interviews and input from the coauthors, Chandre Gould and Garreth Newnam, this section relies on an interview with Melanie Dugmore, the lead consultant in the drafting of the implementation plan for the White Paper on Safety and Security, and Terrence Smith, program manager of the Violence Prevention Program at GIZ that is overseeing the preparation of the implementation plan (expected to be released in 2018). 25 Contents Chapter 6 | Republic of South Africa Systematic Country Diagnostic national objectives. Outcome 3’s objective is that “all people in South Africa are and feel safe” (National Planning Commission 2012: 14). The Justice, Crime Prevention and Security Cluster holds the mandate for coordination of this outcome through implementation of programs by the Department of Justice, the Department of Correctional Services, and the Ministry of Police at the national and provincial level. Following the National Development Plan 2012-2030, multiple government departments at the national level developed policies around various aspects of crime and violence prevention. Because these policies emerged from separate departments with equal authority under the constitution, there is no clear hierarchy. Many of the policies require coordination across departments, but without incentive structures encouraging such partnerships, implementation has been fragmented. And because the plan does not include clear budget lines or coordination mechanisms, it has, in numerous cases, created unfunded mandates. For example, the Integrated Urban Development Framework, developed to define national urban policy, locates community safety as a central issue for urban development and promotes a spatially- differentiated response that accounts for the concentration of violence and crime in urban areas. It argues that many municipal responsibilities affecting safety, such as transportation and emergency services, have been clearly defined and budgeted for in municipal plans, and are institutionalized within strategic planning processes. However, the community safety mandate stemming from the constitution and National Development Plan remains vague and was not costed or resourced, leaving many municipalities struggling to find ways to finance investments that improve safety (COGTA 2016). The framework sits within the Department of Cooperative Governance, which is charged with coordinating across different spheres of government and which is the primary national body with regulatory functions over local governments. The Integrated Social Crime Prevention Strategy, developed by the Department of Social Development (DSD 2011), covers most social aspects of violence prevention. It succeeds the 1996 National Crime Prevention Strategy and the 1998 White Paper on Safety and Security, with the aim of streamlining coordination across government bodies and promoting a deeper focus on prevention by catalyzing a “professional and civil movement, governed and supported by respective departments, which enhances the self-defense capabilities of society” (DSD 2011: 8). The strategy places community involvement at the center of prevention efforts and balances the role of law enforcement with other policy actors. However, because the Department of Social Development was not given clear authority or resourcing to influence other departments, implementation of the strategy has been limited, and it is currently under review. Due in part to the fragmentation of policies, and to the lack of clear roles and authority across different government bodies, implementation of much of the legislation regarding crime and violence has been weak. The failure to cost the legislation or define clear roles and responsibilities across the different departments. As one example, the Domestic Violence Act has been hailed as an example of international best practice. It protects the right of victims to receive “the maximum protection from domestic abuse that the law can provide,” including the issuance of protective orders against perpetrators. The Act also requires SAPS to inform victims of their rights, including the right to shelter. However, the Act does not define which government body is responsible for providing shelters, or place a legal mandate on this provision. In addition, the DVA has not been costed, and as a consequence budget has not been allocated for its full implementation. This has led to inconsistencies 29 26 Republic of South Africa Systematic Country Diagnostic | Chapter 6 in service provision, because resources have been allocated unevenly, for example for police training (DPME 2016). The government of South Africa is in the process of re-evaluating its approach, to strengthen implementation of existing policies, and address the broader underlying drivers of crime and violence. Since 1994, South Africa has endeavored to balance control-oriented approaches focusing on policing with the recognized need to also address broader developmental dimensions of violence. In 2010, the government reviewed the original 1998 White Paper on Safety and Security, and decided to update the policy with two policy interventions: a new White Paper focusing on the policing environment, and a second White Paper on Safety and Security (released in 2016) to provide guidance on an integrated and developmental approach to violence prevention. The 2016 white paper (CSPS 2016) situates law enforcement within a broader approach to crime and violence prevention and attempts to deal with the lack of coherence within the intergovernmental system. It proposes definitions for the roles and responsibilities of various government departments and spheres in relation to community safety. The paper emphasizes the need to build the capacity of local and provincial governments to create enabling environments for community safety. The four pillars of the paper cover the span of interventions, from addressing risk factors to responding to violence once it has occurred: 1 effective criminal justice systems; 2 early interventions, including early childhood development; 3 targeted interventions for at-risk groups; and 4 safety through environmental design. The White Paper proposes the creation of an overarching coordinating body, the National Center on Crime and Violence Prevention, to oversee implementation of national policy. At present, discussion is underway about where the Center would sit, and how it would be financed. The White Paper itself was prepared by the Civilian Secretariat for Police Services, although there is general agreement that it should have a different institutional home. There is discussion about moving its institutional home to Department of Planning, Monitoring, and Evaluation (DPME) where it can (it is argued) more efficiently exercise its monitoring and evaluation responsibilities.¹⁸ An implementation plan for the White Paper is under preparation, which proposes measures to institutionalize the policy mandates put forth in the White Paper. These include clarification of the roles of different bodies, mechanisms for coordination, framework for monitoring and evaluation, and a budget plan for allocation of resources and accountability. The Implementation Plan proposes that the pillars of the White Paper be taken up as focal areas in the Mid-Term Strategic Framework, which ¹⁸ Based on interviews; this will be part of the implementation framework to be released in the next six months. 27 Contents Chapter 6 | Republic of South Africa Systematic Country Diagnostic is the government’s plan of action based on the election cycle, in order to align the activities of all departments and spheres. If approved, the Implementation Plan for the 2016 White Paper on Safety and Security would become the overarching policy for institutionalizing strategies and activities across departments and spheres of government. It is under consultation with different levels of government and civil society. It is set to be released within six months, and will be sent to the cabinet for approval. Also in 2016, the Department of Planning, Monitoring and Evaluation (DPME) undertook a Diagnostic Review of the State Response to Violence Against Women and Children, to assess the institutional and programmatic measures in place to address these types of violence (DPME 2016). The review covers the local, provincial and national level governments, and spans the 11 departments that together share responsibility for preventing and responding to violence against women and children. The Diagnostic Review found that because of the lack of alignment between national policy and provincial and local implementation, prevention activities are not sufficiently prioritized, funded and monitored. Currently, only R9 billion annually is spent on prevention of violence, and this is spread across 11 different national and provincial departments. Overall, spending on prevention amounted to just 2.5 percent of the entire budget of all 11 departments combined (DPME 2016). A separate budget analysis by UNICEF found that the budget for prevention accounted for less than one percent of the combined national and provincial Department of Social Development budgets, and all provinces had seen a decline in allocations toward preventive services, with the exception of Gauteng (UNICEF 2016). Like the 2016 White Paper and corresponding Implementation Plan, the Diagnostic review also emphasized the lack of an adequate coordinating mechanism by which the state could oversee a whole-of-government response, and promote accountability to overarching policy objectives. The specifics of how the Plan of Action resulting from the Diagnostic Review would relate to the Implementation Plan have not been defined, but as the remit of the Department of Social Development, the Plan of Action would fall under the Implementation Plan in the hierarchy of policy strategies. 6.2. Key Institutions, Mandates, and Policies Under South Africa’s decentralized system of governance, the authority for various aspects of crime and violence prevention is spread across multiple spheres. For example, the constitution assigns the competency for policing and law enforcement to the South African Police Service’s national headquarters, which maintains overall command and control functions, while also conferring some power to provincial governments, which in some cases have set up their own oversight mechanisms, such as the police ombudsman office in the Western Cape. Local-level governments represent an autonomous sphere under the constitution, so a local government can establish its own metro police force, its own chief of police, and its own mechanisms for reporting to local oversight bodies. Local governments have ample autonomy in carrying out functions related to crime and violence prevention. While local governments already had a mandate under the Constitution for protecting community safety, the NDP strengthened and clarified this role. The NDP also called for strengthening Community Policing Forums (CPFs) as a means of increasing community participation in public 28 31 Republic of South Africa Systematic Country Diagnostic | Chapter 6 safety.¹⁹ Municipal governments are free to set up new bodies, such as community safety forums, with the approval of their local councils. They are responsible for many urban planning functions, including zoning and the regulation of public spaces, and therefore oversight of the design of safe public spaces. Local governments can address some of the risk factors for violence by enacting bylaws, such as the regulation of informal bars (shebeens). Municipalities and communities are authorized and encouraged to undertake community safety audits to identify assets and strategies and to develop safety plans with corresponding budgets. However, many local governments have struggled carry out their mandates for community safety due to inadequate resourcing. The involvement of local governments in crime prevention programs in South Africa steadily increased after Apartheid in 1994, and some of the most promising examples of effective prevention strategies have been designed and implemented at the municipal and provincial levels. However, the role of municipal governments vis a vis national prevention efforts has not been well-defined, and the distribution of resources has not been aligned to the priority of urban violence prevention (SACN 2017). The government departments responsible for various aspects of violence prevention have varying degrees of autonomy in terms of policy implementation across spheres. The Department of Social Development is responsible for most of the functions relating to the prevention of social crimes, including response to victims, the child protection system, and the overall social welfare system. The national department defines the functions of the provincial departments, but the provincial departments have a certain degree of autonomy over how they perform those functions as they are managed by provincial governments and governed by their legislatures. ¹⁹ The CPFs are a platform established by SAPS in 1995 to convene community stakeholders – including CBOs, NGOs, private sector institutions, youth groups and women’s organizations, among others – with local governments and the police to discuss crime and violence prevention initiatives. They are meant to improve coordination between communities and law enforcement. 29 Contents Chapter 6 | Republic of South Africa Systematic Country Diagnostic The Department of Cooperative Governance and Traditional Affairs (COGTA) is formally responsible for overseeing coordination efforts across spheres and departments. It also exercises some regulatory functions over local governments. The department provides an institutional home for the Integrated Urban Development Framework and is charged with coordinating its implementation. The South African Local Government Association (SALGA) has also taken up the issue of crime and violence prevention, advocating for more support at the local level. SALGA is a voluntary organization with the objective of representing and promoting the interests of local governments vis a vis national and provincial spheres. It hosts a working group on community development that includes a focus on public safety and crime prevention. The National Treasury houses a unit on Neighborhood Development that promotes the sustainable management of urban space. This unit works at the municipal level through a grant mechanism meant to catalyze economic development in particular urban nodes in marginalized areas. The unit is currently looking at how safety can be integrated into the management of public spaces, especially around transit stations or public buildings such as libraries, or public parks. Municipal governments often have several bodies responsible for different aspects of prevention. Community Safety Forums mobilize CSOs, NGOS, the private sector and government for safety planning and dialogue. These are typically done in isolation from provincial governments, though in some cases efforts are underway to better align municipal plans and forums with higher levels of government. Also at the municipal level, Community Policing Forums have been set up to encourage greater interaction of police with communities. Within these CPFs, Youth Crime Prevention Desks recruit youth volunteers (typically 10-15 per precinct) to assist in crime prevention efforts. In some cases municipal governments have clarified the roles of the Desks and linked them into municipal safety planning. GIZ has begun to provide leadership training through the Youth Desks in some areas of the Western Cape that focuses on training youth to assist with safety audits, or mobilizing their peers for different prevention activities. According to interviews with GIZ, job training and entrepreneurial skills have not traditionally formed part of this training – this could be an opportunity for collaboration. The involvement of the private sector in crime and violence prevention has been increasing. One example of this is the creation of the Dialogue Forum on evidence-based programs to prevent violence against women and children. The Dialogue Forum, established in 2015, convenes academics, 33 30 Republic of South Africa Systematic Country Diagnostic | Chapter 6 government officials, private enterprises and representatives of local and international NGOs that evaluate violence prevention programs. The Forum meets twice a year to exchange information on new developments in the field, and develop a shared platform for promoting prevention. Forum partners have collaborated on a series of policy briefs on discrete topics including updates in national policies, recommendations for resourcing violence prevention, and supporting scale-up of programs through multi-sectoral collaboration.²⁰ Private sector forum members are beginning to work with NGOs and international government organizations to find ways to address gender-based violence and promote positive parenting with programs for staff members (ISS 2017a). Key results from the forum include: A better understanding within and across sectors of state budgets and what is required to scale up violence prevention programs; The publication and dissemination of policy briefs authored by government officials, NGOs, and researchers that make the case for violence prevention; Cooperation and collaboration between community-based organizations across geographic divides to improve their sustainability and increase their reach; The creation of a community of people, across sectors, working together to advocate for political support to prevent violence and to raise awareness around the importance of this approach more broadly; and Increased buy-in from government and donors to ensure the sustainability of the forum and to ensure that its work aligns with the needs and priorities of the government and communities. ²⁰ The policy briefs are available at https://issafrica.org/research/policy-brief 31 Contents Chapter 6 | Republic of South Africa Systematic Country Diagnostic 6.3. Spotlight on the Criminal Justice Sector A heavy-handed approach Historically, crime and violence have been viewed primarily as criminal justice issues in South Africa, and the policy response has prioritized law enforcement responses to violence once it has occurred, rather than measures to prevent it. While the international consensus has shifted toward viewing violence as a social and public health challenge to be addressed by a balance of control (policing, criminal justice) as well as measures to address the underlying drivers (WHO 2002, 2013, UN and World Bank 2018), South Africa’s approach remains centered on policing. Law enforcement has consistently taken the lion’s share of budget for the state’s response to crime and violence. Responsibility for prevention is spread across various governmental departments, including social development, health, justice, education, correctional services, police services and others, at both national and provincial levels. Yet most authority and budget is allocated to the SAPS, the Department of Justice and Constitutional Development (DOJCD) and the National Prosecuting Authority (NPA). Over the 2006-2016 period, spending on criminal justice – a sector comprising the police, courts and prisons – rose from R46.6 billion to R126.7 billion, with a 139 percent increase in budget for the police. The growth in spending has resulted in the creation of 68,000 new posts, for a total force of approximately 194,000 police today. Under apartheid, the presence of security forces was disproportionately in white areas in order to enforce the regime, and known for its aggressive approach. By the early 1990s, the South African police had developed an international reputation for brutal, heavy handed tactics and a strong bias against black citizens. The legislation in place at the time tolerated various forms of torture and coercion, and criminal investigations often relied on confessions obtained with harsh interrogation tactics (Rauch 2000). Reforms of SAP (South African Police) began in 1991 with efforts to de-politicize policing (which had been focused on opponents of the apartheid regime) and formed part of the subsequent peace accords. The 1995 South African Police Service Act restructured and reoriented law enforcement efforts. The name of the department was changed from the South African Police to the South African Police Service to reflect a shift from a force to service. It was restructured into four operational levels— national divisions, provincial demarcations, district areas, and stations, with command and oversight mechanisms at each level. National and provincial secretariats for safety and security were created to advise on policy matters and promote civilian control of the police, and community police forums were set up to facilitate greater interaction between local police and communities. In the face of rising crime levels during the transition to democracy, South Africa’s policy increasingly separated policing from other sectors, resulting in a bifurcated approach that increased their heavy- handed nature. While the 1996 National Crime Prevention Strategy had provided space for addressing the social drivers of crime, the South African Police Service adopted increasingly authoritarian policing practices, often assisted by the South African National Defense Force. At the same time, public support for “tough-on-crime” approaches grew, fueling a shift toward increasing arrests and mandatory sentencing (Rauch 2000). 35 32 Republic of South Africa Systematic Country Diagnostic | Chapter 6 Arrests are made in just over half of all cases of reported rape, and a guilty verdict is secured in less than 10% The control-oriented approach emphasized arrests but without the corresponding improvements for streamlining the functioning of the courts or ensuring better coherence across police departments and other parts of the criminal justice sector. For example, the police use different software than the courts, which prohibits data sharing of data and case tracking from arrest through prosecution. The police service’s internal performance management system has arrest quotas, but with no measures to improve the processing of crimes, bottlenecks have arisen. A recent report by the South African Medical Research Council revealed that arrests are made in just over half of all cases of reported rape, and a guilty verdict is secured in less than 10 percent of reported cases (Machisa et al. 2017). Independent sources estimate that the conviction rate for arrests in 2015/16 was 4 percent for crimes such as rape and 24 percent for crimes such as murder. This may be partly explained by the fact that the National Prosecuting Authority has started recommending mediation and other methods of alternative dispute resolution for many cases (ISS 2017b). The focus on control-oriented approaches over prevention has not been accompanied by an overall improvement in the public’s perceptions of police activity. The 2014/15 National Victims of Crime Survey found that 57 percent of households felt satisfied with the police, down from 58.8 percent in 2015 and 64 percent in 2011 (Stats SA 2016). The impact on crime trends is also questionable. While homicides declined significantly—as previously discussed—they appear to be on the rise again. Meanwhile, the private security sector has ballooned, now representing the largest security force in the country per capital and the fourth largest private security industry in the world (Swingler 2017). South Africans spend R45 billion a year on private security, which amounts to one third of what the government spends on the South African Police Service. According to the Private Security Industry Regulatory Authority, there were 488,666 registered private security active officers and 8,692 registered private security companies in South Africa in 2016—more than twice that of the South African Police Service (PSIRA 2014). Larger, especially foreign, businesses are increasingly hiring 33 Contents Chapter 6 | Republic of South Africa Systematic Country Diagnostic private security and reinforcing business premises with additional security features. This includes people employed in security, active guarding, cash-in-transit, and armed response businesses. Impact on Equality and Exclusion The distribution of policing resources has been criticized as uneven, contributing to large inequities in the provision of security services. A 2014 inquiry arising from complaints over the issue resulted in the formation of the Khayelitsha Commission to review police procedures for deploying human resources. The commission found that areas where the majority of residents are poor and black are systematically provided only about one third of the average per capita allocation of law enforcement officers (O’Regan and Pikoli 2014). These are also areas with higher rates of murder and other violent crimes. The commission found that these disproportionate allocations of police have resulted in reduced trust in law enforcement. The police service’s response to the review revealed several barriers to increasing the effectiveness of its efforts. Acknowledging that many stations are perennially understaffed, it explained that the allocation of police officers to particular neighborhoods was based on a formula to discern the “burden of policing.” This formula weighs environmental factors such as inadequate lighting or road access; actual levels of crime is part of the formula, but not as heavily weighted (Redpath and Nagia-Luddy 2015). The police have been criticized for the disproportionate impact of their actions on particular communities. The Victims of Crime Survey for 2015/16 reports that, for all types of crime, coloured households represent the most common victims (8.9 percent) followed by white households (8.3 percent), Indian/Asian households (8.0 percent), and African households (6.9 percent). These differences are small. However, looking at more violent crimes, an analysis of a national sample of 1,378 murder cases conducted by police service in 2009 found that 86.9 percent of murder victims were Africans. Whites made up 1.8 percent of victims, although they comprise 8.85 percent of the population. An analysis by Silber and Geffen (2009) based on data from Statistics South Africa found slightly different numbers but also claimed that, given their share of the population, African and coloured people have been disproportionately represented among murder victims at 81 and 14 percent, respectively, compared with Indian/Asian at 1 percent and white at 4 percent. 37 34 7 CONCLUSION South Africa’s current challenges of crime and violence are rooted in a legacy of exclusion and uneven development. Crime and violence are consistently concentrated amongst excluded geographic and racial groups, where opportunities for socio-economic advancement continue to fall short of real need, resulting in frustrated expectations. While spatial exclusion has decreased under democracy, urbanization continues to outpace the government’s capacity to provide services. As South Africa’s transition continues, there is potential for greater social instability if these long-standing horizontal inequalities are not addressed adequately. Some of the strongest protective factors against violence, in South Africa and elsewhere, are themselves core development goals. Many of the tenets of urban upgrading – promoting stable and secure community environments, and access to safe spaces for recreation – help protect against the risk of violence. At the societal level, addressing historical exclusion through measures to improve infrastructure in traditionally under- served areas, and to improve distribution of economic and educational opportunities can address key risk factors. Thus, many measures to enhance protective factors against violence fit well within the World Bank’s mandate. At the policy level, South Africa appears to be undergoing a transition in its approach to crime and violence. While the foundations for a comprehensive response have been endorsed through national legislative frameworks, implementation has been uneven, and there has been a lack of coherence across the various national bodies responsible for prevention as well as with provincial and municipal levels of 35 Contents Chapter 7 | Republic of South Africa Systematic Country Diagnostic government. At the same time, while legislation has made space for actions to address the social and economic drivers of violence through institutional mandates, these mandates have not been adequately resourced or backed with the appropriate authority for implementation. At present, the processes for developing the Implementation Framework for the 2016 White Paper, as well as the Diagnostic Review and corresponding Improvement Plan, present opportunities for a greater alignment of policy across spheres and more coherence across departments. The establishment of a coordinating body – the National Center on Crime and Violence Prevention – proposed in the White Paper, would constitute another opportunity for better coordinating efforts. Ensuring more coherence across policy bodies depends critically on costing and assigning clear roles. This is a clear recommendation of the various policy reviews, and a key reason behind the current implementation gap. If the White Paper implementation plan can be costed, this will be an important step forward. A second activity could be to conduct an analysis of current spending on crime and violence prevention to identify ways to improve the return on investment in this area. This would be particularly useful at the municipal and provincial levels, where most spending on programs occurs. Important opportunities exist for building on program-level successes. Several municipal-level, area-based urban upgrading projects have integrated violence prevention into their designs, addressing many of these risk factors. These projects are currently poised to be scaled up, presenting an important opportunity for the World Bank to support their rollout in other areas. In addition, there is a growing evidence base for prevention programs that could be scaled up in South Africa. Many address factors identified by the World Health Organization as necessary for the prevention and reduction of violence (Krug et al. 2002). South Africa has reached a critical moment where, if these efforts are to achieve full impact, they need to be scaled up and institutionalized within the various spheres of government. There is a growing national dialogue on prevention that brings together actors from various levels of government, academia, the private sector, and the non-profit sector. The Dialogue Forum, could serve as an important basis for mobilizing efforts at different levels, and the Urban Safety Reference Group presents an opportunity for coordinating efforts in urban areas. Going forward, it will be critical to engage local governments and community organizations at the township level. International best practices consistently indicate that local governments are best equipped to understand risk factors and be able to target them effectively, ultimately ensuring the sustainability of violence reduction. One remaining challenge will be to support leadership within the national government to direct this multi- sectoral and multi-stakeholder agenda. Going forward, the WBG should engage in a comprehensive assessment to identify entry points for further engagement. South Africa’s policy landscape is complex, and the field of violence and crime prevention has become crowded with a variety of actors and pilot initiatives. Any further World Bank consideration of engagement in this area should begin with a multi-level consultation process to seek input from a variety of stakeholders. This should include different spheres of government, communities, academia, the private sector and the nonprofit sector. Because views on the problem of crime and violence, and on the needed measures to address it, will vary across these groups, the most accurate picture will emerge from talking with these different stakeholders. 39 36 APPENDIX Homicide. The killing of an individual. South African criminal law differentiates between murder: “the unlawful and intentional killing of a human being” and culpable homicide: “the unlawful negligent killing of a human being.”²¹ Violence against women. South African legislation does not include a specific definition for violence against women.²² The United Nations defines it as “any act of gender-based violence that results in, or is likely to result in, physical, sexual or psychological harm or suffering to women and girls, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or private life.”²³ Gender-based violence is often conflated with violence against women; however, the two differ in that gender-based violence is understood as violence perpetrated on the basis of one’s gender. This means that men, boys, and people who deviate from gender norms can also be victims. South African legislative frameworks do not contain a specific definition. Intimate partner violence is not specifically defined in legislation, but most research in South Africa uses the World Health Organization’s definition of intimate partner violence as, “behavior within an intimate relationship that causes physical, psychological or sexual harm to those in the relationship.” (Krug et al. 2002).” Violence against children: The South African constitution defines a child as any person under the age of 18. South African legislation does not include a specific definition for violence against children, but the 2016 diagnostic review relies on the World Health Organization’s definition: “All forms of physical, mental violence, injury or abuse, neglect or negligent treatment, maltreatment or exploitation, including sexual abuse” (Krug et al. 2002). Youth violence refers to violence involving people between 12–22 years old, either as victims, perpetrators or both. Nonviolent crime in South Africa is a broad category. The annual Victims of Crime Survey collects data on the following types: car theft, house break-in, home robbery, theft of livestock, theft of crops, theft out of car, destruction of buildings, vandalism, bicycle theft, theft of personal property, hijacking of motor vehicle, robbery, consumer fraud and corruption. These are classified by crimes experienced by individuals, and by households (Stats SA 2018). ²¹ Defined by Supreme Court of Appeal, South Africa, Case 321/2001: Naidoo versus the State. See also Stats SA 2016. ²² See DPME 2016 for a detailed discussion of definitions in the South African context. ²³ UN General Assembly, Declaration on the Elimination of Violence Against Women (DEVAW) 1993. 37 Contents | Republic of South Africa Systematic Country Diagnostic References REFERENCES Adato, M., Carter, M.R. and May, J. 2006. “Exploring Poverty Traps and Social Exclusion in South Africa Using Quantitative and Qualitative Data. Journal of Development Studies 42 (2): 226–47. Artz, L., P. Burton, C. L. Ward, L. Leoschut, J. Phyfer, S. Lloyd, and C. Le Mottee, C. 2016. Sexual Victimisation of Children in South Africa. Final report of the Optimus Foundation Study: South Africa. Cape Town: UBS Optimus Foundation. Beinart, W., P. Delius, and M. Hay. 2017. Land Rights, A Guide to Restitution and Tenure Reform in South Africa. Johannesburg: Jacana. Bhorat, H., A. Lilenstein, J. Monnakgatla, and A. Thornton. (2017). “Socio-Economic Determinants of Crime in South Africa: An Empirical Assessment.” University of Cape Town Working Paper 201704, University of Cape Town. Bruce, D. (2015). Preventing “Crime and Violence Through Work and Wages: The Impact of the Community Work Programme.” South African Crime Quarterly 52, 25–37. Burton, P., and L. Leoschut, 2013. “School Violence in South Africa: Results of the 2012 National School Violence Study.” Centre for Justice and Crime Prevention, Cape Town, South Africa. Caprioli, M., and Boyer, M. 2001. Gender, Violence, and International crisis. Journal of Conflict Resolution. 45: 503-518. Caprioli, M., and Tumbore, P.F. 2003. Ethnic Discrimination and Interstate Violence: Testing the International Impact of Domestic Behavior. Journal of Peace Research, 40: 5-23. Caprioli, M., Hudson, V., McDermott, R., Emmett, C., & Ballif - Spanvill, B. (2007). Putting Women in Their Place. The Bakerournal Of Applied Public Policy, 1(1), 12 - 22. Cassidy, T. et al., (2015). “Evaluation of a Cape Town Safety Intervention as a Model for Good Practice: A Partnership between Researchers, Community and Implementing Agency.” Stability: International Journal of Security and Development 4 (1), art. 27. Cederman, L., Gleditsch, K., & Buhaug, H. 2013. Inequality, Grievances, and Civil War. Cambridge University Press. CJCP (Centre for Justice and Crime Prevention). 2008. National Youth Lifestyle Study. CJCP. Chioda, Laura. 2017. Stop the Violence in Latin America: A Look at Prevention from Cradle to Adulthood. Latin American Development Forum. Washington, DC: World Bank. 38 41 Republic of South Africa Systematic Country Diagnostic | References COGTA (Department of Cooperative Governance and Traditional Affairs). 2016. Integrated Urban Development Framework. http://www.cogta.gov.za/cgta_2016/wp-content/uploads/2016/06/ The-Intergrated-Urban-Development-FrameworkIUDF.pdf Coovadia H., R. Jewkes, P. Barron, D. Sanders, and D. McIntyre. 2009. “The Health and Health System of South Africa: Historical Roots of Current Public Health Challenges.” Lancet 374 (9,692): 817–34. Cluver, L., L. Bowes, and F. Gardner, 2010. “Risk and Protective Factors for Bullying Victimization Among AIDS-affected and Vulnerable Children in South Africa.” Child Abuse Negl. 34 (10): 793–803. Civilian Secretariat for Police Service (CSPS) 2016 White Paper on Safety and Security. Pretoria. DPME (Department of Planning, Monitoring, and Evaluation). 2014. Too Costly to Ignore—The Economic Impact of Gender-based Violence in South Africa. Pretoria: KPMG. ———. 2016. Report on the Diagnostic Review of the State Response to Violence Against Women and Children. Pretoria: DPME and DSD. DSD (Department of Social Development) 2011. “Integrated Social Crime Prevention Strategy.” DSD, Pretoria. DSD, SASSA (South African Social Security Agency), and UNICEF, 2012. “The South African Child Support Grant Impact Assessment: Evidence from a Survey of Children, Adolescents and their Households.” UNICEF South Africa, Pretoria. Duailibi, S. et al. 2007. “The Effect of Restricting Opening Hours on Alcohol-related Violence.” American Journal of Public Health 97: 2,276–28. Edwards, Lawrence, and Asha Sundaram. 2013. “Crime and Firm Dynamics: Evidence from South Africa.” Unpublished manuscript, School of Economics, University of Cape Town. Fang, X., D. A. Fry, G. Ganz, T. Casey, and C. L. Ward, 2016. The Social and Economic Burder of Violence Against Children in South Africa. Pretoria: Save the Children South Africa. http://bit.ly/2Fw4nzJ. Galtung, J. 1969. Peace by Peaceful Means: Peace and Conflict, Development and Civilization. Oslo: PRIO. Herbert, S. 2017. Links between women’s empowerment (or lack of) and outbreaks of violent conflict. Birmingham, UK: University of Birmingham. Hudson, V., Caprioli, M., Ballif-Spanvill, B., McDermott, R., & Emmett, C. (2009). The Heart of the Matter: The Security of Women and the Security of States. International Security, 33(3), 7-45. http://dx.doi.org/10.1162/isec.2009.33.3.7 ISS (Institute for Security Studies). 2003. National Crime and Victimization Survey. ISS, Pretoria. ———. 2016. “Crime Hub.” ISS, Pretoria. 39 Contents | Republic of South Africa Systematic Country Diagnostic References ———. 2017a. “Jet Stores Partners with ISS and Seven Passes Initiative to Expand Primary Violence Prevention Programme.” ISS, Pretoria, October 11. ———. 2017b. “Reducing Violence in South Africa: From Policing to Prevention.” Policy Brief 106, ISS, Pretoria. Jewkes, R., Y. Sikweyiya, R. Morrell, and K. Dunkle. 2011. “Gender Inequitable Masculinity and Sexual Entitlement in Rape Perpetration South Africa: Findings of a Cross-sectional Study.” PLoS One 6 (12): e29,590. Jewkes, R., Y. Sikweyiya, K. Dunkle, and R. Morrell. 2015. “Relationship Between Single and Multiple Perpetrator Rape Perpetration in South Africa: A Comparison of Risk Factors in a Population-based Sample.” BMC Public Health 15 (616, July). Jonck, P., A. Goujon, M. R. Testa, and J. Kandala. 2015. “Education and Crime Engagement in South Africa: A National and Provincial Perspective. International Journal of Educational Development 45, 141–51. Justino, Patricia. 2017. “Linking Inequality and Political Conflict: The Role of Social Mobilization and Collective Action.” Institute for Development Studies. Background Paper for UN-WB Flagship Study, “Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict,” World Bank, Washington DC. Krug, E. G., L. L. Dahlberg, J. A. Mercy, A. B. Zwi, and R. Lozano. 2002. World Report on Violence and Health. Geneva, Switzerland: World Health Organization. La Cava, G., C. Clert, and P. Lytle 2004. “Investing in Youth Empowerment and Inclusion: A Social Development Approach. Insights from the ECA and LAC Regions.” Social Development Paper 60, World Bank, Washington, DC. Langer, A., F. Stewart, and M. Schroyens. 2016. “Horizontal Inequalities and Affirmative Action: An Analysis of Attitudes Towards Redistribution Across Groups in Africa,” WIDER Working Paper Series 119, World Institute for Development Economic Research (UNU-WIDER). Machisa, M., R. Jewkes, C. L. Morna, and K. Rama. 2011. “The War at Home: Gender Based Violence Indicators Project” Gender Links and the South African Medical Research Council, Johannesburg, South Africa. Machisa, Mercilene, Ruxana Jina, Gerard Labuschagne, Lisa Vetten, Lizle Loots, Sheena Swemmer, Bonita Meyersfeld, and Rachel Jewkes. 2017. Rape Justice In South Africa: A Retrospective Study Of The Investigation, Prosecution and Adjudication of Reported Rape Cases from 2012. Pretoria: Gender and Health Research Unit, South African Medical Research Council. Mahofa, G. A. A. Sundaram, L. Edwards. 2016. “Impact of Crime on Firm Entry: Evidence from South Africa.” Economic Research South Africa. National Treasury of South Africa. Pretoria. Accessed 30 July 2018 at https://econrsa.org/system/files/publications/working_papers/working_paper_652.pdf 43 40 Republic of South Africa Systematic Country Diagnostic | References Mathers, C. et al. 2012. The Global Burden of Disease: 2004 Update. Geneva: World Health Organization. Matthews, S., N. Abrahams, R. Jewkes, L. J. Martin, and C. Lombard. 2013. “The Epidemiology of Child Homicides in South Africa.” Bulletin of the World Health Organization 91: 562–68. Matthews, S., R. Govender, G. Lamb, F. Boonzaier, A. Dawes, C. Ward, S. Duma, L. Baraecke, G. Warton, L. Artz, T. Meer, L. Jamieson, and R. Smith. 2016. Towards a More Comprehensive Understanding of the Direct and Indirect Determinants of Violence against Women and Children in South Africa with a View to Enhancing Violence Prevention. Cape Town: University of Cape Town. McCauley, J. et al. 1998. “Inside “Pandora’s Box.” Journal of General Internal Medicine 13 (8): 549–55. Min, E., M. Singh, J. N. Shapiro, and B. Crisman. 2017. “Understanding Risk and Resilience to Violent Conflicts.” Background paper for the United Nations–World Bank Flagship Study, Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict. World Bank, Washington, DC. Mitra, A., and D. Ray. 2014. “Implications of an Economic Theory of Conflict: Hindu-Muslim Violence in India.” Journal of Political Economy 122 (4): 719–65. National Planning Commission. 2012. National Development Plan 2030: Our Future—Make It Work. Department of the Presidency, Republic of South Africa. www.gov.za/issues/national-development- plan-2030. O’Neill, B. C., E. Kriegler, K. Riahi, K. L. Ebi, S. Hallegatte, T. R. Carter, R. Mathur, and D. P. van Vuuren. 2014. “A New Scenario Framework for Climate Change Research: The Concept of Shared Socioeconomic Pathways.” Climate Change 122 (3): 387 400. O’Regan, K., and V. Pikoli. 2014. Towards a Safer Khayelitsha: The Report of the Commission of Inquiry into the Allegations of Police Inefficiency and a Breakdown in Relations between SAPS and the Community in Khayelitsha. Commission of Inquiry into Allegations of Police Inefficiency and a Breakdown. Østby, G. 2008. Inequalities, the Political Environment and Civil Conflict: Evidence from 55 Developing Countries. In F. Stewart, Horizontal Inequalities and Conflict: Understanding Group Violence in Multiethnic Societies (pp. 252-281). London: Palgrave Macmillan UK. Østby, G., Urdal, H., Tadjoeddin, M. Z., Murshed, S. M., & Strand, H. 2011. Population Pressure, Horizontal Inequality and Political Violence: A Disaggregated Study of Indonesian Provinces, 1990- 2003. Journal of Development Studies, 47(3), 377-398. PRIO (Peace Research Institute Oslo). 2017. “Policies for Social and Political Inclusion.” Background paper for United Nations–World Bank Flagship Study, Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict, World Bank, Washington, DC. PSIRA (Private Security Industry Regulatory Authority). 2014. 2013/2014 Annual Report. PSIRA. 41 Contents | Republic of South Africa Systematic Country Diagnostic References Pronyk, P. M. et al. 2006. “Effect of a Structural Intervention for the Prevention of Intimate-partner Violence and HIV in Rural South Africa: A Cluster Randomised Trial.” Lancet 368: 1973–83. Rauch, J. 2000. Police Reform and South Africa’s Transition. Paper presented at the South African Institute for International Affairs Conference. Center for the Study of Violence and Reconciliation. www.csvr.org.za/docs/policing/policereformandsouth.pdf. Redpath, J., and F. Nagia-Luddy. 2015. “Unconscionable and Irrational: SAPS Resource Allocation.” South Africa Crime Quarterly 53. Sánchez, Álvaro I., et al. 2011. “Policies for Alcohol Restriction and Their Association with Interpersonal Violence: A Time-series Analysis of Homicides in Cali, Colombia.” International Journal of Epidemiology 40 (4): 1037–46. SAPS (South African Police Service). 2017. Crime Situation in South Africa, 1 April 2016–31 March 2017. Pretoria: SAPS. Save the Children 2016. “Violence Against Children Costs South Africa R238 Billion.” Save the Children, November 24. www.savethechildren.org.za/news-and-events/news/violence-against-children- costs-south-africa-r238. Silber, G., and N. Giffen. 2009. “Race, Class and Violent Crime in South Africa: Dispelling the Huntley Thesis.” South Africa Crime Quarterly 30. 35–43. www.researchgate.net/publication/284596974_ Race_class_and_violent_crime_in_South_Africa_Dispelling_the_%27Huntley_thesis. SACN (South African Cities Network) 2017. The State of Urban Safety in South Africa Report 2017. A Report of the Urban Safety Reference Group. Johannesburg: South African Cities Network. Stats SA (Statistics South Africa). 2016. Crime Statistics Series Volume III: Exploration of Selected Contact Crimes in South Africa. In‐depth Analysis of Victims of Crime Survey Data: 2011–2014/15. Pretoria: Stats SA. www.statssa.gov.za/publications/Report-03-40-01/Report-03-40-012015.pdf. ———. 2017. “South Africa Demographic and Health Survey 2016. Key Indicators.” Report 03–00– 09. Stats SA, Pretoria. ———. 2018. “Victims of Crime Survey 2016/17.” Stats SA, Pretoria. Stewart, Frances. 2002. Horizontal Inequalities: A Neglected Dimension of Development. QEH Working Paper Series 81. Stewart, F., and A. Langer. 2008. “Horizontal Inequalities: Explaining Persistence and Change.” In Horizontal Inequalities and Conflict, edited by F. Stewart. Horizontal Inequalities and Conflict. Conflict, Inequality and Ethnicity. London: Palgrave Macmillan. Swingler, S. 2017. “South Africa’s R40bn Security Industry Under Threat,” Daily Maverick, May 14. www.dailymaverick.co.za/article/2017-05-14-south-africas-r40bn-private-security-industry- under-threat/#.WUEbo4lEnU. 45 42 Republic of South Africa Systematic Country Diagnostic | References UBS Optimus Foundation 2015. The Optimus Study on Child Abuse, Violence and Neglect in South Africa. Zurich: UBS Optimus Foundation. www.cjcp.org.za/uploads/2/7/8/4/27845461/08_cjcp_ report_2016_d.pdf. UN (United Nations) and World Bank. 2018. Pathways for Peace: Inclusive Approaches to Preventing Violent Conflict. Washington, DC: World Bank. UNICEF 2016. “International Budget Partnership, Budget: Children and South Africa’s Social Development Budget.” UNICEF. UNODC (United Nations Office on Drugs and Crime). 2013. Global Study on Homicide. Vienna: UNODC. USAHRC and UNICEF, 2016. Global Goals for Every Child: Progress and Disparities Among Children in South Africa. Pretoria: UNICEF South Africa. Usdin, S. et al. 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: 2,434–45. Ushakov, D. 2014. Urbanization and Migration as Factors Affecting Global Economic Development. Hershey, PA: IGI Global. Wagenaar, Alexander C., Matthew J. Salois, and Kelli A. Komro. 2009. “Effects of Beverage Alcohol Price and Tax Levels on Drinking: A Meta‐analysis of 1,003 Estimates from 112 Studies.” Addiction 104.2: 179–90. Ward, C. L. 2006. “It Feels Like It’s the End of the World: Cape Town’s Young People Talk About Gangs and Community Violence.” Report to the Institute for Security Studies on the Child Participation Support of the COAV Cities Project. Human Sciences Research Council, Cape Town. Ward, C., E. Martin, C. Theron, and G. B. Distiller. 2007. “Factors Affecting Resilience in Children Exposed to Violence,” South Afr. J. Psychol., 37 (1): 165–87. Western Cape Government. 2014. “Violence Prevention through Urban Upgrading Project.” Western Cape Town government, Cape Town. Willman, A. 2009. “Valuing the Cost of Domestic Violence: A Review by Sector.” In The Costs of Violence, edited by Skaperdas, Soares, Willman, and Miller. Washington, DC: World Bank. World Bank. 2010. Violence in the City: Understanding and Supporting Community Responses to Urban Violence. Washington DC: World Bank. ———. 2015. Doing Business in South Africa 2015. Washington, DC: World Bank. World Bank. 2018. “An Incomplete Transition: Overcoming the Legacy of Exclusion in South Africa.” Systematic Country Diagnostic. World Bank, Washington, DC. https://openknowledge.worldbank. org/handle/10986/29793. 43 References | Republic of South Africa Systematic Country Diagnostic World Health Organization (WHO). 2013. Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-partner Sexual Violence. Geneva: WHO. 47 44