Two-thirds of Johannesburg's residents (62 per cent) (or in some cases members of their households) were victims of crime between 1993 and July 1997.14 The survey results indicate that crime levels in Johannesburg are high, and comparative studies undertaken between 1988 and 1996 show that Johannesburg is not alone. Higher victimisation rates have been documented in the urban areas of most of the developing countries surveyed by the ICVS (Figure 1).15 These general crime rates do not in themselves say anything about the nature of the crime problem in each region, however. It could well be the case that less serious crimes like petty theft constitute the bulk of general crime recorded by the surveys in some countries.

Figure 1: Victimisation levels by all crime in urban areas of selected countries, IC (V)S, 1998 - 1996

Burglary is the most common crime affecting the city's residents.



Although this type of property crime is most prevalent, violent crimes and in particular mugging and robbery, and assault, dominate the victimisation experience.



African residents in Johannesburg are disproportionately affected by violent crime. White and Asian residents are similarly affected by some property crimes.



Most victims of the range of serious crimes covered in the survey have been victims of the same types of crime more than once between 1993 and 1997.



Slightly more than one quarter of all those surveyed have been victims of more than one crime type in the past five years.



Most of these crimes happened to men between the ages of 25 and 60 years.

Contrary to what might be expected in a city such as Johannesburg with its reputation for violence, almost as many property crimes as violent crimes were reported. Of the crimes covered by the survey, 44 per cent were directed against property and 45,5 per cent against people. Since 20 per cent of burglary victims said that violence was used in the course of this crime, fractionally more crimes can be regarded as having been violent.

This trend corresponds with that in other developing countries. Of 13 such countries surveyed by the ICVS, property crimes, such as burglary and car theft, occurred most frequently, followed by serious violent crimes, including robbery and assault. Despite the overall predominance of property crimes, levels of violent crime are nevertheless highest in sub-Saharan African and Latin American countries.16 Although the expectation is that property crimes will be higher in the developed world, studies suggest that in these countries, property crimes have decreased as a result of improved security measures and target hardening (physical measures taken to make committing a crime more difficult). Developing countries are not necessarily less prone to property crime, and instead suffer high levels of both property and violent crime.17

These trends are reflected in Johannesburg. The most common crime experienced by the city's residents between 1993 and 1997 was burglary, with nearly a quarter of the respondents (24 per cent) reporting this crime to the survey. The second most frequently reported incidents were violent crimes: mugging and robbery (16,5 per cent) and assault (15,5 per cent) (Figure 2). As is the case in many developing countries, the vast majority of assaults were of a serious nature, with 84 per cent involving the use of a weapon.

Figure 2: Victimisation levels by property and violent crimes
Johannesburg Victim Survey, 1993 - July 1997


The prevalence of burglary in Johannesburg is not surprising: of 54 countries and cities surveyed by the ICVS, burglary was the crime type occurring most often in Africa.18 After burglary, robbery and assault were the second most prevalent crimes occurring in Johannesburg according to the survey. This is significant, since Johannesburg does not currently feature among the 10 police districts with the highest levels of this crime in the country.19 And since few of these crimes, and in particular mugging, are reported to the police, assault rates are likely to be much higher in Johannesburg than official statistics suggest.

Car theft, which according to other victim surveys in Johannesburg and abroad is the crime most likely to affect city residents, happened to just over 12 per cent of respondents between 1993 and 1997 according to the ISS survey. Although by no means Johannesburg's biggest crime problem, as many as six per cent of respondents experienced violent car theft in the form of car hijacking. The survey did not record details of car ownership, but comparative studies both locally and abroad show that the risk of having a vehicle stolen is twice as high for owners as for the general population. The chances of being a victim of both car theft and hijacking in Johannesburg are therefore probably much higher than the 12 per cent and six per cent respectively recorded by the victim survey. In the 1995 ICVS Johannesburg victim survey, for example, the incidence of car theft was 11,5 per cent for the general population and 24 per cent for vehicle owners.20

The incidence of burglary, robbery, assault and car theft in Johannesburg does not differ markedly from that of urban centres in other countries. Indeed, in some cases, rates of both property and violent crimes are lower here than in certain African, Latin American and Central and Eastern European countries. Of more importance for Johannesburg's residents and for policy-makers, is which people are affected by these crimes.

Probably of the greatest concern is the fact that most victims surveyed in Johannesburg have been the targets of the same type of criminal activity more than once. This was the trend for all major crimes covered in the survey, except car hijacking (Figure 3). It is cause for concern, since multiple victims are less inclined to report these repeat crimes. This tendency also heightens the fear of crime, as well as people's vulnerability to further attacks.21 Equally unsettling is that 28 per cent of all those interviewed had experienced more than one type of crime in the past five years.

Figure 3: Multiple victimisation
Johannesburg Victim Survey


Crimes which are aimed at property affect those people and those parts of Johannesburg which present the greatest opportunities for theft: according to the victim survey, the white and Asian communities (which are generally wealthier than the African and coloured communities), as well as people living in the suburbs are most at risk (Figures 4 and 5). Most Johannesburg residents (over the age of 15 years) are equally at risk of robbery and mugging, although Asians, residents of the inner city and the elderly are slightly more vulnerable. In the case of burglary and car theft, white and Asian people are disproportionately victimised compared to the proportion of the population they represent (Figures 4 and 6). Just over half (52 per cent) of car theft victims reported that the crimes happened near their homes, the majority of which are in the suburbs. It is unlikely that police patrols can effectively prevent these crimes in the vast suburban areas of Johannesburg. More viable solutions are target hardening and other community-based preventive measures.

Figure 4: The victims of crime
Johannesburg Victim Survey



Figure 5: Victims' places of residence
Johannesburg Victim Survey



Figure 6: Racial composition of the survey sample
Johannesburg Victim Survey


The use of violence in the course of these property crimes is more of a threat for African victims than for other groups. Violence was used in 20 per cent of all burglaries reported to the survey. One quarter of African victims experienced violence, compared with only nine per cent of white burglary victims. Similarly, between a quarter and one third of victims living in townships, informal settlements and, to a lesser extent the inner city, reported the use of violence, while this was the case for only 14 per cent of victims living in the suburbs.

The greater risk of violence for Africans in the course of property-related crimes is also evidenced by car hijacking trends. As many as 73 per cent of hijacking victims were African (Figure 7). Contrary to popular media representations which depict the victims of car hijacking as wealthy and white, Africans in Johannesburg are most at risk. Since fewer Africans probably own cars than whites, if car ownership is taken into consideration, a relatively small group of Africans comprise the high risk group for hijacking. This trend may reflect lifestyle patterns, such as where and when people travel, as well as the nature of precautions taken.

Figure 7: Car hijacking victims
Johannesburg Victim Survey


It is worth noting, however, that respondents were not asked whether they were driving the vehicle that was hijacked, or what the purpose of their journey was. In some cases, hijack victims may well have been commercial drivers of delivery vans or other vehicles. The fact that 70 per cent of these crimes did not occur in the victims' area of residence but elsewhere in Johannesburg, could support this argument. And although 33 per cent of victims live in the suburbs and 20 per cent in the inner city (Figure 8), these proportions are not far off the distribution of the areas of residence for all survey respondents (Figure 9).

Figure 8: Car hijacking victims' places of residence
Johannesburg Victim Survey



Figure 9: All survey respondents' places of residence
Johannesburg Victim Survey


Africans living and working in Johannesburg are not only more exposed to violence in the course of property crimes, but also to violent interpersonal crimes such as assault, murder and rape. More than three quarters of the victims of assault, rape and murder in Johannesburg were African (Figure 10) and in the case of murder and rape, the majority of victims live(d) in townships. According to the police in Johannesburg, many of these violent crimes (including rape) are associated with alcohol consumption, and victims and offenders are often known to one another.22 This is borne out to some extent by the survey results (Figure 11). Thirty per cent of assault and 44 per cent of rape victims said they knew the offender, at least by sight. It is difficult to estimate the extent to which these crimes refer to domestic violence. Incidents are unlikely to be reported as domestic violence to a general crime survey, and the fact that most violent crimes did not occur in the victim's home, is perhaps indicative of this (Figure 12).

Figure 10: The victims of violent crime
Johannesburg Victim Survey



Figure 11: Whether victims knew offenders
Johannesburg Victim Survey




Figure 12: Where violent crimes were committed
Johannesburg Victime Survey



According to the survey results, men are most at risk of violent crime in Johannesburg. This is not always the trend internationally, and probably relates to the fact that sexual incidents and domestic violence the kinds of violent crimes women are often vulnerable to are unlikely to be reported to a street victim survey. In western countries (such as in Europe and North America), studies show that men and women are equally at risk of assault. In the rest of the world, however, women are more likely to be assaulted than men, and indications are that in many developing countries, the problem of violence is largely one of sexual and/or non-sexual violence against women.23

Although the extent of sexual crimes against women was poorly captured by the victim survey, data from alternative sources in Johannesburg verify some of the tentative survey findings. During 1997, in a study conducted by the Johannesburg Sexual Offences Forum, the socio-economic profiles, as well as details about rape incidents, were recorded for 786 victims reporting to district surgeon's offices in hospitals and clinics in parts of Greater Johannesburg.24 Well over 70 per cent of these victims were African (as reflected in the crime survey), with the majority aged between 13 and 30 years. Of the 42 per cent of victims who knew their attackers, most of these were known by sight only, although as many as 19 per cent of the known offenders were either relatives or partners.

Although rape is less likely to be reported to the authorities when the offender is known to the victim, the proportion of rapes committed by strangers is nevertheless large in Johannesburg. The circumstances in which these crimes occurred present opportunities for preventive action by local authorities, the police and potential victims. Almost one quarter of rapes occurred in open ground and most were probably related to commuting patterns and evening social activities, since rapes peaked between 18h00 and 22h00. Perhaps the most startling finding given that most victims did not know their offenders was that the most common location at which reported rapes occurred was in the rapist's home (30 per cent)25 and that 64 per cent of rapes involved abductions, most of which lasted for up to 12 hours. This suggests the extent to which offenders disregard the ability of the police to apprehend and convict them for these crimes.

The perception of impunity from prosecution among offenders is probably one of the reasons why Africans are more at risk of violent crime in Johannesburg. Access to police and other criminal justice resources have in the past been skewed in favour of white South Africans. Police resources remain unevenly spread, with well over half of police stations still located in formerly white areas. Other hypotheses have sought to link institutional violence by the apartheid government and the political violence which developed in the fight against apartheid in the 1980s, to high levels of domestic and interpersonal violence. This allegedly resulted in the economic, social and political emasculation of men, encouraging them to reassert their position through controlling and directing violence towards vulnerable groups such as women and children.26 Also understood as the `cradle of violence', violence at home is thought to encourage the use of violence in general.27

Socio-economic explanations point to links between urbanisation, economic strain, rapid population growth rates, lifestyle patterns and levels of education on the one hand, and high rates of violence on the other. These factors explained 57 per cent of the variance in victimisation rates in 49 countries (that participated in the ICVS study). Violent `contact' crimes were closely related to economic deprivation, particularly when directed against women.28

Internationally, the second most important factor in explaining high levels of violence is gun ownership. In developing countries and countries in transition such as in Latin America and South Africa where many people own handguns, the risk of violent crime is particularly high. This is also the case in the United States. Gun ownership in these regions has been related to high levels of robbery, assault, and sexual crimes.29 In South Africa, uncontrolled and high levels of alcohol consumption in an environment of poverty have also been associated with crime, and in particular violent crime.30

The fact that a range of factors are related to high levels of violent crime supports the argument that economic growth alone will not be enough to reduce crime, since this both alleviates economic strain and presents new opportunities for crime. In places where, for example, the possession and use of firearms have become rife and the sale and consumption of alcohol are uncontrolled, special policies need to be developed to address these problems.

In countries in transition from authoritarian to democratic rule, other factors are equally relevant. It is likely that the use of violence also relates to a breakdown in law enforcement, as well as to increased target hardening in the context of economic deprivation and a failing criminal justice system. (The installation of sophisticated electronic devices to prevent car theft and bank robberies, for example, have led to violent car hijackings and the use of other violent tactics in the course of crimes which often remain undetected).