INTRODUCTION

It is often said that South Africa has the highest levels of crime in the world and that Johannesburg is the world’s crime capital. How true is this? South Africa’s crime situation is usually compared to that in developed countries as many developing countries do not keep proper and reliable records on the crimes that are committed there. This is, however, an unfair comparison. Socio-economic conditions, unemployment levels and the state’s ability to help those in need are more favourable in the developed states concentrated mainly in the northern hemisphere. Moreover, developed states tend to have sufficient resources to pay for a well-resourced criminal justice system, and to increase spending as and when an upsurge in crime occurs.

South Africa is a developing country. According to the 1999 Human Development Report, South Africa occupies position 101 out of 174 in terms of human development (where Canada, at position 1, scores the highest in terms of human development, and Sierra Leone at position 174, the lowest).¹ According to the Human Development Index, South Africa scores the highest among sub-Saharan countries. For example, Namibia is at position 115, followed by Botswana (122), Zimbabwe (130), Kenya (136), and Tanzania (156).

Given South Africa’s position in Africa and the fact that it is largely a developing country, the question which should be asked is: How does South Africa’s levels of crime compare to other developing countries — especially those in sub-Saharan Africa?

It is a difficult question to answer as crime statistics for many African states are notoriously unreliable or non-existent. The two most complete sources for African crime statistics come from the International Criminal Police Organisation (Interpol), which provides figures on the actual number of crimes reported to the police in various countries, and victim surveys which have been conducted in a selected number of African countries by the United Nations.

SOUTH AFRICA COMPARED

By global standards, South Africa has extraordinary high levels of violent crime, according to 1997 Interpol statistics.² In 1997, South Africa had the highest per capita rates of murder and rape, the second highest rate of robbery and violent theft (after the Bahamas), and the fourth highest rates of serious assault and sexual offences, of the 110 countries with crime levels listed by Interpol.³

South Africa’s per capita levels of property-related crime, commercial crime and drug-related offences were surpassed by a number of other countries, including numerous developed countries. South Africa’s property crime levels compare favourably with those of many developed countries if measured on a per capita basis. However, a better indication of property crime is to measure it in relation to the number of units of property.

For example, in 1997 some 248 cars were stolen per 100 000 people in South Africa. In France, 552 cars were stolen per 100 000 of the population. There are, however, about seven times as many cars on the roads of France than in South Africa (and probably even more if the cars from other European countries which travel through France every year are taken into account). This means that the chance of a car being stolen in South Africa is twice as high compared to France. Measured in this way, South Africa’s vehicle theft levels are not as favourable as Interpol’s figures indicate.

The 1997 Interpol report provides crime data for nine sub-Saharan countries only (including South Africa). The per capita crime rate for the major violent crimes of murder, rape, and robbery and violent theft was substantially higher in South Africa than in the other eight sub-Saharan countries. South Africa also had the highest per capita serious assault rate (541 per 100 000 of the population), but this was closely followed by Namibia (497 per 100 000).

The Interpol figures also indicate that, of the nine listed sub-Saharan countries, South Africa had the highest per capita levels of theft, in general, and vehicle theft specifically. Care needs to be taken when comparing South Africa’s vehicle theft figures with that of other sub-Saharan countries, however. According to the Interpol figures, the per capita level of vehicle theft in South Africa (248 per 100 000 of the population) was almost four times as high as in the second worst affected country, Swaziland (70 per 100 000). Given South Africa’s relative wealth and economic dominance in the region, it is likely that there are substantially more cars in South Africa than in other sub-Saharan countries.

In respect of burglary, Swaziland had the highest level in 1997 (810 cases per 100 000 of the population), followed by South Africa (780 cases per 100 000), and Namibia (580 cases per 100 000).

VICTIMS OF CRIME SURVEY

Police crime statistics are generally considered to be the most accessible source of information on how many crimes are committed in a country. However, police records contain only information on crimes which are reported to and recorded or uncovered by the police. For a number of reasons, such information is not sufficient. Because of inadequate recording practices and political agendas, crime statistics sometimes reflect only the police’s performance in recording crimes. Moreover, some crimes are reported but not recorded. This happens when victims call the police and the police judge the crime as not sufficiently serious to warrant its recording.⁴

Crime victims do not report crimes for a number of reasons.⁵ The propensity to report crimes is likely to differ from country to country and by type of crime. Generally, for serious property-related crimes - especially where the stolen property is insured - the reporting rate is high. However, for some violent offences - especially sexual offences - reporting rates tend to be low.

The general tendency is for people to report the more serious property crimes.

Victim surveys seek to overcome the phenomenon of underreporting of crime and the fact that the police do not record all crimes which are brought to their attention. Ideally, victim survey results should be read in conjunction with official crime statistics:

In 1996-97, the United Nations Interregional Crime and Justice Research Institute (UNICRI) conducted victim surveys in 17 countries of the developing world.⁸ Seven of these countries are in Africa, six in Latin America and four in Asia.⁹ The respondents were generally drawn from the population of the largest cities in the selected countries. Respondents were asked about their victimisation experiences over a one-year period preceding the survey.

Overall, the selected Asian countries consistently ranked the lowest for all types of crime, while African and Latin American countries had the highest victimisation rates for all types of crime. All regions were mostly affected by theft (7% in Asia and 13% in Africa and Latin America over a one-year period), while victimisation rates for all other types of crime were lower.¹⁰

VIOLENT CRIME

According to the 1996-97 UNICRI victim survey, for the crime type of sexual assault (which includes rape, attempted rape and indecent assault, but excludes offensive behaviour), Ugandans indicated the highest level of victimisation of 4.5% over a one-year period. This was followed by Egyptians (3.1%) and South Africans (2.3%). Victimisation levels in respect of sexual assault were the lowest in Botswana (0.8%). African levels of sexual assault compared favourably with some Latin American countries. For example, Brazil had 8% and Colombia 5%. Asian levels, however, were the lowest.

Robbery was the highest in Tunisia (6.3%), followed by Zimbabwe (4.8%) and South Africa (4.6%).¹¹ As with sexual assault, Botswana had the lowest victimisation level for robbery (2%) of the six African countries surveyed. Some Latin American victimisation levels were again higher for robbery: Brazil had 11.3% and Colombia 10.6%.

‘Assault with force’ victimisation levels were the highest in Zimbabwe (6.7%) followed by South Africa (5.6%) and Botswana (3.7%).¹²

For the ‘assault with force’ category, most of the surveyed African countries registered higher victimisation rates than Latin American countries, with surveyed Asian countries registering the lowest victimisation levels.

Sometimes, the offenders used a weapon to commit robberies, assaults and sexual assaults.¹³ Overall, weapons were used more frequently in the developing world than in the other regions surveyed for the commission of these crimes.

Among the African countries surveyed, South African victims were the most likely to state that the offender used a weapon in the sexual assault (37.1% of sexual assaults), and robbery (73.3%). Tanzania had the highest incidence of assault where a weapon was used (70.4%) followed by South Africa (62.9%). In all other African countries surveyed, less than half of the assaults involved the use of a weapon.¹⁴

Victims of assault were asked whether they suffered personal injuries as a result of the aggression perpetrated against them. The seriousness of the injuries can be gauged by the need of victims to ask for a doctor’s assistance after the incident.

In Tanzania, 53.9% of assault victims were injured and 32% saw a doctor. This was followed by South Africa where 43% of assault victims were injured and 31.7% consulted a doctor. In all other African countries surveyed, the injury rate was lower as was the propensity of assault victims to see a doctor after the incident. The injury rate was also generally lower in the Asian and Latin American countries surveyed compared to the African countries.¹⁵

PROPERTY CRIMES

According to the 1996-97 UNICRI victim survey, Tanzanians experienced the highest levels of victimisation with regard to burglary and attempted burglary. Thus, 19% of the Tanzanian survey respondents stated that they had been a victim of a burglary, and 12.9% said that they had fallen victim of an attempted burglary over a one-year period.

According to the survey, South Africa had one of the lowest risks of burglary and attempted burglary compared to the other surveyed African countries: 6.3% and 3.7% of South Africans stated that they had been the victim of, respectively, a burglary or attempted burglary over a one-year period.

Actual theft during burglaries happened more frequently in developing countries than in the rest of the world. In poorer areas, stolen goods typically include money, food and inexpensive household items such as clothing or linen, which are most probably stolen for the personal use of the burglar.

In wealthier areas where homeowners generally keep their money in a bank and other valuable items in safes, burglars tend to take what is available and give preference to objects which are easily resold. In this respect, the most frequently stolen objects are those that are easier to place on the market of stolen goods, such as electrical appliances, television sets and radios.

"The difference in objects stolen between affluent and poor regions is opportunity determined, in terms of the type of objects available, or the level of protection. Furthermore, the level of protection determines the damage involved in breaking into the house, which is also an indicator of the degree of difficulty that the burglar has to face in committing the offence. It is therefore clear that behind burglaries in well or poorly protected houses, there will be different cost-effectiveness calculations that will take into account whether the prospective target is worth the risk involved in committing the crime."¹⁶

Of all the African countries surveyed, South African burglary victims were the most likely to state that something had been stolen (94.9% of burglaries), or that damage had been caused (62.6% of burglaries). In fact, in this respect South Africa experienced the highest level of damages occurring during the course of a burglary of all 17 developing countries surveyed. In respect of objects being stolen during the course of a burglary, South Africa was placed second after India.

The UNICRI survey also asked vehicle owners whether any of their household vehicles (including trucks and vans) had been stolen (vehicles taken away for the purposes of ‘joyriding’ were included in the question).

South Africans indicated the highest level of vehicle thefts with 7.6% of surveyed vehicle owners indicating that one of their household vehicles had been stolen. This was followed by Uganda (5.9%) and Tanzania (5.8%).

In respect of theft from a vehicle (such as, for example, a car radio or money left lying inside a vehicle), Tanzanians indicated that they were the worst affected with just over a quarter of vehicle owners indicating that something has been stolen from one of their household vehicles over a one-year period. This was followed by Zimbabwe (21.2%) and Tunisia (17.8%). Among South African respondents, only 14% of vehicle owners stated that one of their household vehicles had something stolen from it.

An important indicator of the nature of vehicle theft and of the efficiency of the police’s response to this type of crime is the data on vehicle recovery rates. It can be argued that a high recovery rate indicates that vehicles are stolen for joy-riding purposes. This might explain why vehicle recovery rates are very high in industrialised countries (84% in North America and 74% in Western Europe). "The recovery rate in the developing world is much lower than in the industrialised world which, comparatively speaking, indicates both lower levels of efficiency in recovery and less joy-riding."¹⁷

Among the African countries surveyed Zimbabwe had the lowest recovery rate (30.2% of stolen vehicles recovered), followed by South Africa (45.8%). In all the other surveyed African countries, more than half of the stolen vehicles were recovered.

OTHER CRIMES

Apart from the types of crime which have been dealt with so far, other forms of victimisation covered by the UNICRI victim survey included vehicle vandalism, theft of a motorcycle, theft of a bicycle, theft of personal property and assault without force. During the year preceding the interviews, approximately a quarter of the respondents from all participating countries experienced at least one of these types of victimisation.

If all these crimes are taken together, Zimbabweans experienced the highest level of victimisation (32.7%) of the seven African countries surveyed. This was followed by Uganda (28.9%) and Tanzania (28.2%). South Africa was in the middle with 19.8%. South Africa experienced relatively high levels of assault without force (5.2%), second only to Zimbabwe (8.7%).

Taking the category of ‘other crimes’ together, the surveyed African countries compared poorly to the Asian countries which generally registered considerably lower victimisation rates. A number of Latin American countries, however, had the highest victimisation rates of all the countries surveyed. For example, Colombia had the highest with 35.1% followed by Argentina (30.7%).¹⁸

FEAR OF CRIME

The UNICRI survey asked respondents whether they felt very safe, fairly safe, a bit unsafe or very unsafe in the street after dark. Of all the African countries surveyed (and all Latin American and Asian countries), South Africans were the most likely group to say that they felt very unsafe in the street after dark (39.6%). After Botswana (7.4%), South Africa was the country where the fewest respondents felt very safe (14.3%).

A further question gauging people’s fear of crime was whether they avoided places when going out after dark. Among the Africans countries surveyed, Zimbabwe had the highest proportion of respondents stating that they avoided places when going out after dark (56.7%), followed by Egypt (54.4%) and South Africa (51.7%).

A third question seeking to measure the fear of crime asked respondents whether they felt that a burglary was likely to occur in their household in the next year. Respondents in Botswana were most likely to state that a burglary was very likely in their household in the next 12 months (26.6%), followed by South Africans (25.1%), and Tanzanians (23.3%).¹⁹

POLICING LEVELS

It is sobering to realise that levels of policing in a country have a limited impact on crime levels only. Generally, developed states have a favourable police to population ratio. For example, in Italy the ratio is 1:102. That is, for every uniformed police officer or detective there are 102 people living in Italy. Portugal has a ratio of 1:225 and Germany of 1:315.

In South Africa, there is one police officer for every 408 people. However, some developed countries have a police to population ratio which is not as favourable as South Africa’s. For example, in Canada the ratio is 1:552 and in New Zealand, 1:557, yet the levels of serious violent crime in these countries are significantly lower than in South Africa.

However, compared to more developed countries, South Africa’s civilian to police officer ratio does not reveal the full picture. While the South African Police Service (SAPS) is large, with 105 000 uniformed police officers and detectives, the service’s effectiveness is hampered by a number of factors:²⁰

South Africa has a high crime rate, especially in respect of serious and violent crime. The investigation of such crimes consume considerable police resources.



South Africa has a relatively youthful population. In 1996, over 44% of the country’s population was under the age of 20.²¹ Young people commit more crime than adults - especially violent crime.²²



South Africa is geographically large with the result that police resources are thinly spread in the country’s rural areas.



South Africa has a rapidly urbanising population. The anonymity provided by expanding cities and the unstructured terrain of the growing number of informal settlements make effective policing difficult.



Many officers in the SAPS are poorly trained. According to the previous national police commissioner, George Fivaz, a quarter of all police officers are ‘functionally illiterate’. Moreover, 34 800 have only ten years of schooling or less, and over 8 000 do not have a driver’s licence.²³

South Africa’s police to population ratio compares favourably to that of many developing states — especially sub-Saharan countries. While South Africa has a police to population ratio of 1:408, Tanzania has a ratio of 1:998 and Malawi of 1: 1 830.²⁴

CONCLUSION

In terms of crimes reported to the police, South Africa has extraordinary high levels of violent crime — both by global and by African standards. Moreover, in respect of property crime — especially vehicle theft — South Africa has one of the highest levels of crime among sub-Saharan states for which figures are available.

However, when victim survey results are analysed, this picture changes. In respect of violent crime South Africa’s crime levels are high but not the highest when compared to a number of other African countries. It is striking, however, that South African victims of violent crime are more likely to be attacked or threatened with a weapon compared to the victims of all the surveyed African countries. Not surprisingly, South African assault victims are very likely to suffer serious injuries.

In respect of property crimes, South Africans fall somewhere in the middle in relation to other African countries surveyed by UNICRI. However, in respect of burglary, South African victims were the most likely to suffer some loss through burglary, while South African vehicle theft victims are — after their Zimbabwean counterparts — the least likely to have their vehicle recovered by the police.

It is undeniable that levels of crime in South Africa are high by global and African levels — especially in respect of violent crime. It would seem, however, that in respect of a number of violent crimes and most property crimes, South Africans are less likely to be victimised than many Africans inhabiting a number of countries on the rest of the continent.

NOTES

UNDP, Human Development Report 1999, United Nations Development Programme, Oxford University Press, New York, 1999, pp 164-167.



International crime statistics 1997, International Criminal Police Organisation (Interpol), Lyon, in The incidence of serious crime January to June 1999, Semester Report 2/99, Crime Information Analysis Centre, November 1999, Pretoria, annexure D. The 1997 Interpol figures were the latest available figures at the time of writing. Since 1997, the level of reported crime has increased in South Africa. Between 1997 and 1999, the number of reported crimes in South Africa increased by 12%. The number of reported violent crimes increased by 13% over the same period.



The per capita rate of sexual offences (which include the crime of rape) was higher than South Africa’s in the Bahamas, Dominica and Saint Kitts and Nevis. The per capita rate for serious assaults was higher than South Africa’s in Granada, Australia and Lithuania.



A Alvazzi del Frate, Victims of crime in the developing world, Publication 57, United Nations Interregional Crime and Justice Research Institute, Rome, 1998, p 6.



See A Louw, Crime in Pretoria. Results of a city victims survey, Institute for Security Studies, Halfway House, August 1998, pp 18-24.



Alvazzi del Frate, op cit, p 7.



J J M van Dijk, The state of crime and criminal justice worldwide, paper read at the 10th United Nations Congress on the Prevention of Crime and the Treatment of Offenders, 2000, p 2.



Alvazzi del Frate, op cit.



The countries are the following. Africa: Botswana, Egypt, South Africa, Tanzania, Tunisia, Uganda and Zimbabwe; Latin America: Argentina, Bolivia, Brazil, Colombia, Costa Rica and Paraguay; Asia: China, India, Indonesia and the Philippines.



Alvazzi del Frate, op cit, p 29.



UNICRI defines robbery as "a violent crime aimed at stealing property from the victim."



Alvazzi del Frate, op cit, pp 34-36.



The International Crime Victim Survey referred to any type of weapon or object used as a weapon. The text of the question addressed to victims reads: "Did (any of) the offender(s) have a knife, a gun, another weapon or something used as a weapon?" (see ibid).



Ibid, pp 73-74.



Ibid, pp 65-66.



Ibid, pp 63-64.



Ibid, p 61.



Ibid, pp 40-42.



Ibid, pp 115-121.



In April 2000, the SAPS was employing 124 160 people. Of these, 105 560 were employed as uniformed police officers and detectives, while the remainder (18 600) were civilian staff.



Statistics South Africa (SSA), Census in brief: The people of South Africa population census, 1996, Report no. 03-01-11, SSA, Pretoria, 1999, p 22.



M Schönteich, The dangers of youth? Linking offenders, victims and age, Nedcor ISS Crime Index 3(5), September — October 1999, pp 22-28.



Written answer to a parliamentary question (National Assembly) by the minister for Safety and Security, question no 309, Internal question paper no 12, 10 March 2000.



Letter from Interpol, Pretoria, 4 June 1999.