Focus on violent crime welcome but likely to lead to fiddling of figures
A single-minded focus on getting the numbers of certain crimes down obscures the truth and makes for ineffective policing solutions
● In his state of the nation address last week, President Cyril Ramaphosa said South Africans should “work together to ensure that violent crime is at least halved over the next decade”.
There can’t be many South Africans who don’t support Ramaphosa in this aspiration. It is also a simple matter to point out the potential obstacles, from factors of an economic and social nature to the institutional and political, that may stand in the way of such a target being achieved.
But there is another, more obvious potential pitfall. Any focus on reducing violent crime will inevitably translate into pressure on the police to show that it is going down. The risks of such pressure are well illustrated by SA’s experience during Thabo Mbeki’s second term of office as president, which started in 2004.
That year, the government and police leadership agreed to set a target. During the coming years, they said, they would endeavour to reduce violent crime by 7%-10% a year. In the five years to 2009 they appeared to come within reach of this target. SA achieved a reduction of 25% in seven of the major violent crime categories. (Due to changes in the classification of sexual offences it is not possible to consider them in this article.)
Closer scrutiny of the statistics highlights certain apparent contradictions. During the five financial years from 2003/2004 to 2008/2009, murder declined from 19,824 cases to 18,148, a modest drop of 8.5%. During the same period, however, attempted murder declined by almost 40%, from slightly over 30,000 to close to 18,300.
Dramatic disparities were also recorded with aggravated robbery, which is officially classified into seven subcategories. Three of these are the so-called trio crimes: home robbery, business robbery and car hijacking. These are the crimes that, in the discourse of the rich, came to be known as the “crimes most feared by South Africans”. The idea that these crimes are the most feared is pure myth and contradicted by survey data on the topic. Nevertheless, the myth has an enormous impact. Among subcategories of aggravated robbery, these crimes enjoy special status in the priorities of the South African Police Service (SAPS) and National Prosecuting Authority, a status that is shared only by cash-in-transit heists.
During the five-year period up to 2008/2009, the recorded number of trio crimes increased by 44%, from 26,821 to 38,544. Bank robberies, cash-in-transit heists and truck hijackings also all increased, by 56%, from 1,147 to 1,784. Miraculously, though, overall levels of aggravated robbery still dropped.
Levels of recorded street robbery dropped precipitously by 26% from 105,690 to 77,984 cases. Street robbery is the biggest category of aggravated robbery. Despite the fact that six of the seven subcategories of aggravated robbery had increased, declines in recorded street robbery enabled the SAPS to record an 11% reduction in aggravated robbery.
A similar decline, of 27%, was also recorded in the two categories of assault. In combination, cases of assault with intent to inflict grievous bodily harm and common assault dropped from 541,024 to 396,615, a decrease of 144,000. The biggest drop in any single crime category was common assault — there were 88,104 fewer cases during the five-year period.
It is obvious that what was happening had nothing to do with reduced levels of attempted murder, street robbery or assault. In the categories of violent crime that the police, criminal justice system, politicians and the middle-class public have always prioritised, recorded violent crime was either increasing or, in the case of murder, decreasing at a modest rate.
In these categories, the accurate recording of reported crime obviously continued to be important. It was therefore in the categories of violent crime regarded as less significant that recorded violent crime was plummeting. This had nothing to do with effective policing. The principal person driving the reduction in figures was Jackie Selebi. The pressure he placed on his commanders to bring the violent crime figures down filtered down the chain of command. Media stories about the failure to record crimes showed that it was taking place in at least five provinces. At one police station, for instance, officers were instructed not to record criminal complaints unless the suspects were immediately available or the facts easily ascertainable. Unrecorded dockets were allegedly kept separate and eventually burnt.
The phenomenon is not unique to SA. Harvard academic Malcolm Sparrow said in a 2015 paper: “Relentless pressure to reduce the number of crimes reported, without equivalent pressure to preserve the integrity of the recording and reporting systems, invites manipulation of statistics. The most obvious forms of manipulation involve suppression of reports (failing to take reports of crime from victims) and misclassification of crimes to lower categories in order to make the serious crime statistics look better.”
This then raises a conundrum in terms of our response to violent crime. In order to tackle it, we have to abandon the fixation with getting the numbers down. Instead, an institutional environment that emphasises the integrity of the system for recording crime reports must be created in the police. This requires not only professionalising the police, but a change in the mindset of their political overseers.
Instead of a single-minded focus on crime statistics, police responses to violent crime need to be based on a disaggregated analysis of violent crime. Violent crime is not one thing — it is usually dispute-related violence, sexual violence or robbery (sometimes it is a combination of these).
In his speech, Ramaphosa stated that violent crime “is a societal problem” and “requires a society-wide response”. This is correct. The police cannot fix many of the underlying problems that contribute to violent crime. Nevertheless, they need to play a pivotal role in efforts to reduce violent crime. These efforts need to be based on evidence-based analysis of each of the major forms of violent crime, and strategies that are clearly aligned with this evidence.
Evaluation of these strategies can’t simply rely on crime statistics. It needs to be able to assess whether data on levels of crime is reliable. Only if there are indicators that the police are encouraging reporting and being rigorous about recording it, can their statistics on violent crime be used to assess if strategies to reduce violence are working or not.
Bruce is an independent researcher specialising in crime and policing