Crime in South African

Measuring crime, particularly over sustained periods, is fraught with difficulties. Recorded crime levels undercount the real levels of crime as they do not reflect unrecorded crimes. For crime to make it onto the official police records two things need to happen. Firstly victims or witnesses must report it to the police. Secondly the police must record the crime in their records.

According to Statistics South Africa’s 1997 national Victims of Crime Survey, crimes involving valuable and insured property are mostly reported. For example, 95% of vehicle thefts, 60% of vehicle hijackings and 59% of burglaries are reported. Less serious property crimes and interpersonal violent crimes are more often not reported than reported. Thus, only 41% of robberies, 38% of assaults and 28% of thefts of personal property are reported.¹

During the first four years after South Africa’s political transition in 1994, overall crime levels almost stabilised, albeit at very high levels of especially violent crime. Between 1994 and 1997, recorded crime increased at an average of only 1% per year. Thereafter levels of recorded crime, measured from one year to the next, increased at an escalating rate. Overall crime levels increased by almost 5% between 1997–98, 7% in 1998-99, and 7.6% in 1999–2000 (Table 1).²

Table 1: Percentage change in the number of crimes recorded,
for two four-year periods between 1994 and 2000

The latest available crime statistics at the time of writing are those for the period January–September 2001. During the first nine months of 2001, 1,844,000 crimes were recorded by the South African Police Service (SAPS); up from 1,464,000 over the same nine-month period in 1994. If the January–September period for 2001 is compared with that of 1994, the number of recorded crimes increased by 26%.³

While recorded crime increased between 2000 and 2001, the rate of increase is slowing down. If the first nine months of 1998 are compared with 1997, recorded crime increased by 4%. Thereafter, recorded crime increased by 7% (1998/99) and 8% (1999/00). The January–September 2001 period experienced a 2% increase compared to the same period in 2000 – the lowest year-on-year increase since 1996/97. While the figures indicate a slowing down in the rate of increase of recorded crime in 2000/01, it needs to be remembered that this is occurring at a point where recorded levels of violent crime are extraordinarily high. The time period is too short, moreover, to draw a firm conclusion whether recorded crime might be entering a declining phase, or whether the first nine months of 2001 reflect a period of stabilisation (as was the case during 1996), after which recorded crime will continue to escalate.

The crime rate (as measured per 100,000 of the population) increased from 5,173 crimes in 1994, to 5,635 crimes in 2000.⁴ At the 2000 level, the total risk of being a victim of crime per person per year is 5.6%, even before unrecorded crimes are considered.

While murder levels declined after 1994, overall levels of violent crime have experienced the greatest increase compared to all other crime categories. Between 1994 and 2000, violent crime increased by 34%, property crime by 23%, violent crime against property (i.e., arson and malicious injury to property) by 10%, commercial crime by 9%, and drug- and drunken driving-related offences by 1%. When measured over the 1999–2000 period, the same trend emerges, with violent crime increasing at the greatest rate (Figure 1).

Figure 1: Percentage change in the number of crimes recorded in South Africa, 1994 - 2000 and 1999 - 2000

Consistently high levels of violent crime – and the extensive media coverage of it – have resulted in a significant increase in the public’s feelings of insecurity. Annual Human Sciences Research Council (HSRC) public opinion surveys ask a nationally representative sample of respondents about their feelings of personal safety. In 1994, almost 75% of respondents said they felt safe, while less than 20% felt unsafe. At the end of 2000, respondents were almost equally divided with 44% feeling safe and 45% feeling unsafe.⁵ (The HSRC’s 2001 survey does not include a question on feelings of personal safety.)

Crime in Port Elizabeth

Recorded crime rates differ significantly between provinces. In 2000, the highest provincial per capita rates of serious crime were recorded in Gauteng, and in the Western and Northern Cape. Overall, the Eastern Cape fared well, coming sixth or lower in the crime rankings out of the nine provinces, with the exception of stock-theft (3rd position) and murder (5th).

Crime levels in the country’s metropolitan areas tend to be higher than in the country as a whole. Most factors associated with high crime rates characterise cities to a greater extent than they do small towns. Population density, for example, is thought to be associated with crime, in that greater concentrations of people lead to competition for limited resources, greater stress and increased conflict. Other factors that characterise urbanisation, such as overcrowding and high levels of gang activity, are mainly evident in urban areas and are known to be related to criminal activity.⁶

Recorded crime levels vary between cities.⁷ On the basis of recorded crime figures for 2000, Johannesburg has significantly higher levels of crime than other large cities. In 2000 just over 18,300 crimes were recorded per 100,000 residents of the Johannesburg police area, compared to 8,361 for Port Elizabeth. During 2000, people living in Johannesburg were over two times as likely to be victims of crime as those in Port Elizabeth (Figure 2).

Figure 2: Recorded crime rate in selected urban police areas, 2000

In respect of a wide range of recorded serious crimes, Port Elizabeth is one of the safer metropolitan areas in the country. This can partly be attributed to police initiatives that successfully target high-crime areas in Port Elizabeth, and inter-departmental co-operation in the form of successful Integrated Justice System projects.

In 2000 there were 105 recorded murders for every 100,000 residents of the Johannesburg police area, followed by Cape Town, Durban and Port Elizabeth (all 72), and Pretoria (35). During 2000 the national recorded murder rate was 49 per 100,000.

In respect of robbery with aggravating circumstances (i.e. robbery involving a dangerous weapon), Johannesburg was also at the top of the list for 2000, with 1,565 robberies per 100,000 of the population, followed by Pretoria (449), Durban (440), Cape Town (378) and Port Elizabeth (322).

A number of violent crimes are called “social fabric crimes” by the SAPS, because many of these offences are committed by people known to one another in familiar environments. It is in respect of the social fabric crimes of rape and serious assault where Port Elizabeth fares poorly. In 2000 Port Elizabeth had the highest per capita recorded rate of rape and assault with the intent to inflict grievous bodily harm (assault GBH), compared to other large South African cities (Figure 3).

Figure 3: Recorded rape and assault GBH in selected urban police areas, 2000

Rape and assault statistics should be treated with caution as many cases are not reported. The figures are nevertheless interesting because they differ markedly from those of other crimes. The consistency in the ranking of areas for rape and assault GBH suggest that the nature of the two crimes are related. Many of these offences undoubtedly occur in the domestic and social context between people who know one another, or in the context of criminal gang activity.

Compared to national crime trends, Port Elizabeth appears to enjoy relatively low increases in the levels of recorded crime. The number of crimes recorded in the Port Elizabeth police area increased from 63,500 in 1994, to 67,600 in 2000 – an increase of just over 6% over a seven year period. This compares favourably with the national trend. In South Africa the number of recorded crimes increased by 24% between 1994 and 2000. During 2000, about one-third of the crimes recorded in Port Elizabeth involved violence or the threat of violence, while half were property-related crimes.

While overall recorded crime levels increased moderately in Port Elizabeth, some crimes decreased. Between 1994 and 2000, murder, assault GBH, theft out of motor vehicle, and commercial crime decreased. An even greater number of serious crimes declined in Port Elizabeth between 1999 and 2000 (Table 2).

Table 2: Percentage change in selected crime in the Port Elizabeth police area

Drug-related crime increased significantly between 1999 and 2000 in Port Elizabeth. This is largely attributable to intensified police action to crack down on those possessing and dealing in drugs. Drug-related crime, like drunken driving offences, relies primarily on police action for its detection. An increase in these types of crime is consequently a positive indicator of police performance.

Performance of the criminal justice system

South Africa’s criminal justice system is not performing optimally.⁸ In 2000, in the region of 2.58m crimes were recorded by the police. Of these, approximately 610,000 (24%) went to court, and the prosecution service took 271,000 (11%) cases to trial. These resulted in slightly more than 211,000 (8%) convictions, the other cases being withdrawn or settled otherwise.⁹

In other words, out of the initial 2.58 million cases recorded, 8% resulted in the conviction of the perpetrators. For some serious crimes the number of convictions as a proportion of recorded cases was even lower. For car hijacking it was 2%, aggravated robbery 3%, arson 4%, residential burglary 5%, and rape 7.6%. To phrase it differently, in 2000 only one out of forty-three car hijackers whose crimes were recorded, were convicted and punished for their crimes.

It is difficult to identify reliable and fair performance indicators for the criminal justice system. It is the purpose of the criminal justice system to combat, prevent and reduce crime. In accomplishing this purpose the performance of the criminal justice system is influenced by a variety of issues – such as prevalent societal values, unemployment levels, and the proportion of young men in a population – factors over which it has no control.

Similarly, it is difficult to identify accurate and equitable performance indicators for one part of the criminal justice system. The performance of an individual part of the criminal justice system (such as the police, prosecution, or prison service) is also influenced by outside factors, over which the individual part or the system as a whole has little or no control. One such factor, for instance, is the willingness of members of the public to report crime to the police.

The performance of one section of the criminal justice system is also influenced by other parts of the system of which it forms a vital part. For example, a widely used performance measurement for the prosecution service is the number of recorded crimes that result in a conviction. However, the strength of the prosecution’s case and its ability to secure a conviction depends not only on the skills of the prosecutor, but also on the proficiency with which the case is investigated by the detective service, and the reliability and honesty of the state witnesses. A prosecutor has only limited control over such factors. However, a good and diligent prosecutor will exploit the limited control that he has and – to use the present example – guide and assist the detective in the investigation of the case, and consult thoroughly with witnesses. The conviction rate thus remains a useful performance measurement tool for the prosecution service, provided the aforementioned limitations are taken into account.

It is consequently possible to develop performance indicators for individual parts of the criminal justice system, as long as the limitations of the indicators are recognised, and the indicators are seen holistically as reflecting the performance of an integrated and inter-dependent system.

Two important indicators can be identified by which the performance of the criminal justice system can be measured. Firstly, the number of cases that the police solve sufficiently well for the prosecution to decide to take on the case in court. Secondly, the number of cases that result in a successful prosecution. The South African criminal justice system is performing poorly on the basis of the first indicator, but well on the second.

Conviction rate and declining prosecutions

Of the 271,060 cases the prosecution service took on in 2000, convictions were obtained in 211,760 cases. That is, once cases are successfully processed through most of the criminal justice system, with the suspects ending up in the accused box in court, their chances of being convicted are a high 78%. Unlike the police, however, prosecutors have the luxury of being able to decide which cases to take on. Generally, the prosecution service elects to proceed in the prosecution of a suspect only where “there is sufficient and admissible evidence to provide a reasonable prospect of a successful prosecution”.¹⁰

There is a considerable variation in the conviction rate between types of crime, however. During 2000, over 75% of common assault, residential housebreaking and stock-theft prosecutions resulted in a conviction, while prosecution rates for arson and rape were below 60% (Figure 4).

Figure 4: Cases resulting in a conviction as a proportion of cases prosecuted in South Africa, 2000

While the prosecution service is successful in convicting most of the offenders it prosecutes, the number of cases taken on by the service declined at a time when recorded crime was increasing. In 1994/95, 350,200 prosecutions and 260,900 convictions took place. This decreased to some 271,100 prosecutions and 211,800 convictions in 2000. The number of serious crimes, as recorded by the SAPS, increased by 481,000 between 1994 and 2000.¹¹ Between 1999 and 2001 there has, however, been an increase in the number of prosecutions and convictions.

In other words, while the number of recorded serious crimes increased by 24% between 1994 and 2000, the number of prosecutions dropped by 23% and convictions by 19%. The chances of the average offender being caught and punished consequently declined between 1994 and 2000.

Case withdrawals

In 2000, 46% of the cases referred to court by the police were withdrawn in court. This is not surprising in cases where the victim and the offender are known, or even related, to each other and where the victim might decline to testify against the accused, nor in cases where the trial might be unreasonably delayed due to outstanding investigations (e.g. a district surgeon’s report). Unsurprisingly, therefore, as a proportion of the number of cases sent to court by the police, about half of rape and assault cases are withdrawn in court.

However, about half of all residential housebreaking, robbery, and car theft cases that were referred to court in 2000 were also withdrawn by the prosecution service. For these crimes it is unlikely that the victims are the reason for the high number of withdrawals. The more likely reasons are inordinate delays in the investigation of these crimes, and the failure of witnesses to testify in court. The latter might be because witnesses are intimidated from attending court by the criminals they are supposed to testify against. Many burglaries, robberies and car thefts are committed by crime syndicates who do not hesitate to intimidate those who might testify against them. Moreover, some witnesses might have no faith in the criminal justice system, and elect not to testify for this reason.¹²

During 2001, 423,890 cases were withdrawn by the prosecution service.¹³ Of the withdrawn cases, 92% were withdrawn at district court level, and 8% at regional court level. Thus, of all the cases dealt with by the prosecution service during 2001, just over half were withdrawn. A further 37% resulted in a conviction and 9% in an acquittal.

Since 1996 there has been a steady increase in the number of cases withdrawn by the prosecution service. The marked increase in the number of withdrawn cases after 1999 is partly the result of an initiative started in mid-2000 by the National Prosecuting Authority (NPA), to encourage prosecutors to withdraw cases which are flawed in some way and not ready to proceed to a successful prosecution.¹⁴

The increase in case withdrawals has limited the impact Operation Crackdown has had on prosecutions and convictions. Between 2000 and 2001 the number of cases referred to court increased by 30%. Over the same period the number of case withdrawals increased by 44%, while the number of prosecutions and convictions increased by only 16% and 15% respectively.

Outstanding and finalised cases

Since April 1999 the National Prosecuting Authority’s Court Management Unit has collected data on the number of finalised and outstanding criminal court cases. This is too short a period to enable a reliable identification of performance trends. Nevertheless the figures give a good indication of the magnitude of the case backlog the prosecution service is facing.

Between April 1999 and July 2001, the country’s regional courts (which deal with the bulk of all serious criminal trials) finalised an average of 3,010 cases a month, but had an average of 43,500 cases per month outstanding on the courts’ rolls. The actual number of cases finalised per month by the regional courts increased over the 28-month period, but the number of new cases coming into the system increased at an even greater rate. As a result more cases were outstanding on the regional courts’ rolls in July 2001 than in April 1999. As part of a January 2000 performance agreement, regional courts should finalise an average of 15 cases per month.¹⁵ By July 2001 the average regional court was finalising just less than 10 cases a month, some five cases per month below target.

According to the National Prosecuting Authority, 184,253 uncompleted criminal cases were outstanding at the end of 2000 and were carried over to 2001. A further 756,801 new criminal cases entered the magistrates’ or lower court system during 2001.¹⁸ No figures are given for the high courts, but less than 5% of all criminal cases end up there.¹⁹ Of the new cases for 2001, 55,178 (or 7%) went to the regional courts where most serious crimes are tried, and 701,623 (93%) went to the district courts where less serious offences are prosecuted.

In 2001, 358,123 cases were finalised with a verdict: 81% resulted in a conviction and 19% in an acquittal, or not guilty, finding. Of the cases finalised with a verdict about nine-tenths (88%) were finalised in the district courts and 12% in the regional courts. The conviction rate was higher in the district courts (83%), compared to the regional courts (66%).²⁰

Between 2000 and 2001 there was a marked increase in the number of cases referred to court (Figure 5).²¹ This is the likely result of the police’s new operational approach, also known as “Operation Crackdown”. The operation is a high-density, zero tolerance-type police and army operation, and has taken place in high crime areas since April 2000. Some 462,000 arrests were made during the first twelve months of the three-year operation.²²

Figure 5: Cases processed by the prosecurtion service, 1996 - 2001

The 16% increase in the number of prosecutions between 2000 and 2001, was less than the 24% increase between 1999 and 2000. Between 2000 and 2001 the number of regional court prosecutions went up by 33% and the number of district court prosecutions by 14% (Figure 6).

Figure 6: Number of cases finalised with a verdict in district and regional magistrates' courts in South Africa, 1999 - 2001

At the end of 2001, 181,638 lower court cases were outstanding or had not been finalised (Figure 7). This represents the backlog of cases the courts had to contend with at the beginning of 2002. The backlog is high given that the number of outstanding cases at the end of 2001 was equal to just over half of all cases prosecuted during that year.

Figure 7: Number of lower court cases outstanding nationally at end of year, 1999 - 2001

In October 2000 the National Director of Public Prosecutions, Bulelani Ngcuka, pointed out that the 180,000 criminal cases outstanding in the country’s courts at the time, would take prosecutors two years to deal with, excluding any new cases.²³

Compared to 2000, the number of outstanding cases decreased by 1% during 2001 – a significant achievement given that the number of new cases referred to court increased by 30% over the same period. This decline in outstanding cases was achieved by a 20% reduction in outstanding regional court cases. (Outstanding district court cases actually increased by 6%).

The decline in outstanding cases during 2001 would not have been possible without the approximately 150 Saturday courts and additional courts that were operational during that year. These courts disposed of almost 15,000 cases between February and December 2001 – 62% of them regional court cases.²⁴

While Saturday and additional courts contributed to the finalisation of a substantial number of cases, the productivity of the average magistrate’s court declined. During 2001 the average district court secured 130 convictions (down from 137 in 2000) and the average regional court 114 convictions (down from 135 in 2000). Given that district courts mainly deal with minor offences, 130 convictions per average court per year – or a conviction every other working day – is low.

Regional courts deal with serious crimes such as murder, rape and armed robbery. Trials in the regional courts typically involve a number of state witnesses and the presentation of non-oral evidence such as blood samples and district surgeons’ reports. As a result many regional court trials last a number of days. Given the slow rate at which regional courts are able to finalise cases, it will take these courts some time to reduce the backlog of cases.

Detention cycle time

Some accused are incarcerated while they await the outcome of their trial. This is because the courts refuse to grant them bail, or because bail is set at an amount that is unaffordable to the accused. There are a number of factors that determine the length of time an accused spends in prison awaiting the finalisation of his or her trial. The speed with which the police finalise the investigation, the length of the trial, and the number of postponements the accused requests during the trial are all factors which lengthen the awaiting trial period of an accused, and over which the prosecution has little or no control. However, in many cases there are delays in the finalisation of trials because the courts’ rolls are badly managed, something over which the prosecution has some influence.

The detention cycle time, or the average length of time unsentenced prisoners remain incarcerated until the finalisation of their trials, increased during 2001; from 136 custody days in December 2000 to 145 custody days in December 2001 (Figure 8). In mid-1996 the detention cycle time was a relatively low 76 days.

Figure 8: Average number of days unsentenced prisoners incarcerated nationally, 1999 - 2001

This means that on average, at the end of 2001, accused persons were imprisoned for four-and-a-half months awaiting the finalisation of their trial. Such delays in the processing of cases involving awaiting trial prisoners place a considerable financial burden on the department of correctional services. In 2001 a prisoner cost the department approximately R88 a day. Multiplied over an average of 145 custody days, this comes to almost R13,000 per average unsentenced prisoner.

The average detention cycle time varies significantly between the provinces. In February 2002 provinces with above average detention cycle times were the Limpopo Province (208 days), Gauteng (157 days) and Western Cape (155 days). The Northern Cape had the lowest detention cycle time of
92 days, with the Eastern Cape in the third lowest position (108 days).

Prison overcrowding

The number of unsentenced prisoners increased from 19,571 in June 1994 to 55,285 in December 2001 – a massive increase of 183%. Over the same period the number of sentenced prisoners increased from 79,987 to 120,005 – an increase of 50% (Figure 9).

Figure 9: Changes in prisoner numbers, 1994 - 2001

Between May 2000 and June 2001, the number of unsentenced prisoners decreased by 16% from 61,590 to 51,559. This decrease is largely attributable to the release, in September 2000, of some 8,200 prisoners who could not afford to pay bail amounts of R1,000 or less and who had been charged with non-violent offences. The National Prosecuting Authority also encourages prosecutors to withdraw badly investigated cases and to work longer hours, in an attempt to reduce the number of unsentenced prisoners.²⁵

Notwithstanding the short-term decrease in the number of unsentenced prisoners, South Africa’s prisons are overcrowded. In December 2001 the country’s 239 prisons were holding 175,290 inmates but had an approved occupancy level of only 105,435. In other words, the national prison occupancy level was 166%. The province with the highest level of overcrowding was Limpopo Province with an occupancy level of 249%. The Eastern Cape’s prisons had an occupancy level of 175% – the fourth highest of the country’s nine provinces (Figure 10).

Figure 10: Prison occupancy levels (%) by province, December 2001

High levels of overcrowding place considerable strain on the human and capital resources of the department of correctional services. This impedes the department’s ability to manage its prisons and rehabilitate prisoners. According to the department, overcrowding has an “adverse effect on offenders, staff and the safe custody of prisoners”.²⁶ Overcrowding exacerbates tension, hostility and aggression between prisoners, and between prisoners and prison personnel. During 2000/01, 2,361 assaults by prisoners on prisoners were recorded by the department (up from 2,271 in 1999/2000), and 619 assaults by prison personnel on prisoners (up from 559).²⁷

Speaking in February 2001, the minister of correctional services, Ben Skosana, highlighted the problem the department is facing with the spread of HIV/AIDS in prison. According to Minister Skosana, “overpopulation, which leads to over-utilisation of sanitation facilities, contributes to poor standards of cleanliness and sanitary conditions in some prisons, and the spread of communicable and contagious diseases”.²⁸ The prison environment tends to exacerbate factors that lead to the spread of HIV/AIDS. High-risk sexual behaviour (such as unprotected sex), sexual abuse, rape, gang violence, the sharing of skin piercing instruments for tattooing, and the high turnover of offenders which inhibits the treatment of sexually transmitted diseases, all lead to a higher transmission rate of HIV/AIDS among prisoners. The number of natural deaths in prison have increased from 186 in 1995, to 1,087 in 2000 – a massive increase of almost 500%.²⁹

A potential further consequence of overcrowding is an increase in escapes. To the department’s credit, the number of persons escaping from correctional services’ facilities is declining. In 2001, there were 227 escapes, down from 254 in 2000 and 1,244 in 1996. However, because of the lack of prison space, an increasing number of prisoners (primarily awaiting trial) are held in SAPS holding cells. During 1996, some 1.4m people passed through police holding cells. In 2000 this had increased to 1.6m. One result of this increase has been a rise in the number of people escaping from police holding cells, from 3,595 in 1996 to 4,797 in 2000. Consequently, the total number of prisoners who escape from prisons and police holding cells has increased since 1996 (Figure 11).

Figure 11: Number of prisoners who have escaped from prison and SAPS holding cells nationally, 1996 - 2000

Given the above performance indicators, it is apparent that the South African criminal justice system is unable to act effectively as a deterrent against crimes that can be policed, and against criminals who could be deterred. The day-to-day activities of the criminal justice system, such as the investigation of crime, the prosecution of offenders, and the processing of criminals through the system, are occurring at low performance levels.