Crime stats – the devil is in the detail

This article was written on 25 October 2017, following the release of the 2016/2017 SAPS National Crime Statistics on 24 October 2017. 

Yesterday the Minister of Police released the 2016/17 crime statistics. As always the ‘good news’ aspects of reported crime in South Africa were immediately emphasised, namely that the number of cases of reported ‘serious crime’ had declined by 1.8%. However, ‘serious crime’ includes 17 sub-categories of crime, and over the past 20 years there have been numerous allegations by crime researchers and analysts of substantial and persistent under-reporting for many of these sub-categories, especially the various forms of robbery, assault, and sexual crimes.

Perception surveys have suggested that this under-reporting has been due to: a lack of trust in the police; the inaccessibility of police stations (particularly in rural areas); adverse notions of the utility of reporting criminal incidents to the SAPS; concerns about being stigmatised by family and/or community members; and fears about being further victimised by the perpetrator of the original crime (especially in the case of domestic violence and rape). Hence, lumping all serious crimes together as an aggregate indicator of criminal offending is unhelpful and misleading.

The aggregate manner in which the crime data and statistics are presented also masks the disproportionate nature of violent crime in South Africa, which for decades has been concentrated in about 20 per cent of police precincts. Most of these high crime precincts are densely populated, infrastructurally marginalised and characterised by elevated levels of poverty. In addition, high crime areas are potentially perilous for those SAPS members that are required to work within them. That is, these spaces are forbidding and difficult to navigate often due to a lack of streets, systematic dwelling numbering systems and lighting. Police have been frequently at risk of being attacked by armed criminals. Community confidence in the police is often severely lacking. Therefore, in seeking to gauge a more informed analysis of crime a focus on high crime precincts provides a more enlightening touchstone for the state of violent crime in South Africa.

Globally murder (or homicide) is widely regarded as one of the most statistically reliable crime categories. The reason for this is that the occurrence of this intentional form of killing is most often linked to a corpse, with such a corpse typically being examined by a medical official who is required to declare the apparent manner of death. This information is then logged in the death registry and a death certificate is issued. This process has commonly been followed in South Africa. Given the problem of reliability of other reported crime categories, murder is used in many countries as a crude measure of the state of violent crime in general.

Drilling down to precinct level, murder data for the high crime areas makes disquieting reading. In many areas traditionally affected by gang violence in the Western Cape, the murder levels increased by more than 20% compared to 2015/16 reporting year. This included Delft (28% increase), Bishop Lavis (26% increase) and Philippi East (23% increase). In Bishop Lavis and Philippi East the annual number of murders has in fact doubled over the past 10 years. In other high crime areas in the Western Cape there were even more dramatic increases in murder, such as Worcester (55% increase) and Paarl East (77% increase).

Similarly, in KwaZulu-Natal there have been alarming escalations in murder cases in Kwadukuza (50% increase) and Plessislaer (37% increase). Umlazi and Inanda have seen increases in the region of 20% increases, and Chatsworth experienced a 17% increase. In Gauteng there have been between 20% and 30% increases in murder in Evaton, Orange Farms, Rietgat, Booysens and Dobsonville; whilst both Ivory Park and Roodepoort experienced more than a 50% increase in reported murder cases. In the Eastern Cape there were more than 15% increases in murder in Dutywa and Port St Johns, more than 26% increases in Libode and New Brighton.

There were however significant decreases in murder in some traditionally high crime areas, such as Gugulethu (26% reduction), Empangeni (26% reduction), Tembisa (16% reduction), Hillbrow (13% reduction) and Mitchells Plain (13% reduction).

Murders in South Africa are typically the outcome of aggressive verbal disagreements between young men in the context of alcohol consumption, or during the perpetration of another crime, such as robbery. In some areas, gang rivalries result in killings. Research on masculinity and violence suggests that such lethal confrontations have been closely tied to notions of manhood that reveres dominant and aggressive men. Such a type of masculinity encourages bravado, weapon-carrying, excessive drinking of alcohol and dramatically increases the risk of violent confrontations between men and groups of men in public spaces. The link between this form of masculinity and murder is amplified when men at risk of perpetrating violence live in communities where there are generally high levels of interpersonal violence; and where firearms are easily obtainable.

Consequently the role of the police in reducing and preventing murder and other violent crime is somewhat limited. Indeed, as the Minister of Police has emphasised, a more professional and service-oriented SAPS is an essential component, combined with police actions that result in the confiscation of illegal firearms, as well as the arrest and successful conviction of perpetrators. However, what is ultimately required is a whole-of-society and whole-of-government approach to reducing and preventing violent crime. Such an approach is actually included in the recent White Paper on Safety and Security that has been championed by the Civilian Secretariat for Police Services. Therefore a priority for government should be the full-scale implementation of the recommendations included in this key policy document.

Dr Guy Lamb is the Director of the Safety and Violence Initiative at the University of Cape Town.