SA CRIME QUARTERLY No 8 JUNE 2004 17 Duxita Mistry, Institute for Security Studies duxita@iss.org.za FALLING CRIME, RISING FEAR 2003 National Victims of Crime Survey For several years, the police have maintained that crime levels in South Africa are ‘stabilising’. Without alternative sources of crime statistics, it is impossible to test these claims. The most reliable supplements to police data are national victim surveys, which are now conducted regularly in several countries for precisely this purpose. The 2003 National Victims of Crime Survey shows that crime levels, as measured by the surveys, have indeed declined since 1998. Public sentiment does not reflect this good news however – feelings of safety are much worse now than they were five years ago. I n 2003, the Institute for Security Studies (ISS) conducted a national victim survey with the aim of measuring crime trends in the country, public perceptions about crime and safety, as well as confidence in the criminal justice system. The study was planned and carried out to allow direct comparisons with the national survey conducted in 1998 by Statistics SA for the Department of Safety and Security and the United Nations Interregional Crime and Justice Research Institute (UNICRI). The survey was conducted between September and October 2003. Households were randomly selected across the country based on the 2001 Census, and a national sample of 4,860 people, over the age of 16 years, was realised. The sample was stratified by province and urban/rural areas, and the data was weighted to reflect the actual composition of the population. In a nutshell, the findings revealed a drop in crime levels since 1998, although accompanied by rising, and high, levels of public insecurity. In particular, people were most afraid of violent crimes such as murder and sexual assault. This article provides an overview of some of the results, and explores the relationship between fear of crime and actual experiences of victimisation.1 Why the need for a national victim survey? National victim surveys provide invaluable information on victimisation rates and vulnerable groups, because they focus on the victims of crime (rather than the perpetrators as is the case with police and court data), and cover a representative sample of the population in a specific geographic area. The surveys also provide an understanding of public perceptions of crime and safety, and the fear of crime, as well as victims’ actual experiences of specific types of crime. They also offer insight into the underreporting of crime, or the so-called ‘dark figure’ that describes the incidents that do not make it into police records. As such, victim surveys complement police statistics by adding to the information that is already on the official database. Victim surveys have a distinct advantage in that they show the extent of multiple victimisation and whether crimes are concentrated in a small number of people who are frequent victims, or are spread out among the general population. Disadvantages of victim surveys are that respondents do not always Table 1: % of South Africans, over the age of 16 years, who were victims of crime in 1998 and 2003 1998 2003 Any crime 24.5 22.9 Housebreaking 7.2 7.5 Corruption* - 5.6 Theft of personal property 4.8 4.7 Stock theft 4.9 2.5 Theft out of vehicle 2.5 2.5 Assault 4.2 2.2 Robbery 2.4 2.0 Deliberate damage to vehicle 1.3 1.3 Bicycle theft* - 1.2 Car theft 1.2 1.0 Deliberate damage to buildings 1.1 0.9 Fraud 3.0 0.8 Crop theft* - 0.7 Car hijacking** 1.4 0.5 Other crime 1.6 0.2 Murder 0.5 0.2 Theft of motorbike 0.0 0.1 Sexual assault/rape 0.4 0.1 SA CRIME QUARTERLY No 8 JUNE 200418 MISTRY some extent. A more dramatic decrease is evident for crimes like stock theft, assault, and fraud. The levels of sexual offences/rape as measured in both surveys should be treated with caution as these crimes are underreported in victim surveys as well as official police statistics. The apparent decline in sexual offences from 0.4% in 1998 to 0.1% in 2003 is therefore unlikely to reflect the real trend, as data on these crimes is not considered reliable in the survey context. recall precise details of their experiences, the results are subject to sample error, and the surveys are a poor means of collecting data on crimes such as rape, domestic violence, fraud and corruption, because members of the public are often reluctant to discuss such matters with survey fieldworkers. General victim surveys also do not record crimes against businesses, crimes against children, and drug and firearm related offences. Nevertheless, the value of these studies has been recognised by governments in several developed countries such as the United Kingdom and the United States, where victim surveys are now conducted annually to supplement police statistics. Together with the official crime data, the survey results are key for crime prevention, policing and justice policy formulation, for identifying gaps in resource allocation, as well as improving victim support services. The good news: crime rates down since 1998 In order to gain the maximum benefit from national victim surveys, similar studies must be conducted at regular intervals. A comparison between the 2003 and 1998 surveys shows whether or not victimisation rates have increased, and how levels of fear of crime may have changed over time. In the latest survey, just more than one fifth (22.9%) of all South Africans had been a victim of crime in the 12 months preceding the survey. This is slightly less than the overall victimisation rate recorded by the 1998 survey, in which one quarter (24.5%) of South Africans had experienced crime over the preceding year. This means that overall, the victimisation rate dropped by 1.6% over the past five years. It follows that most of the crime types measured in the surveys would also show a decrease between 1998 and 2003. The one exception to this trend is housebreaking, which is the only category of crime in the survey that increased, albeit fractionally, since 1998 (Table 1). Rates of theft out of motor vehicles and deliberate damage to motor vehicles remained the same, while other offences such as theft of personal property, car theft, deliberate damage to buildings and robbery have decreased to * crime types not covered in the 1998 survey ** In the 1998 survey the category ‘car hijackings’ included attempted and ‘successful’ hijackings, while in the 2003 survey only successful hijackings were recorded. This could account for the decrease in the hijacking rate between 1998 and 2003 reflected here. they did in 1998, if the “very safe” and “fairly safe” categories are added together (Figure 1). However, significantly more felt only fairly safe in 2003 as opposed to very safe in 1998. The tendency towards feeling less safe becomes a clear trend when the night-time results are considered. South Africans felt significantly less safe when walking alone after dark in 2003 than they did five years ago (Figure 2). In fact, more than double the number of people in 2003 than in 1998 felt very unsafe walking in their area after dark (58% in 2003 as opposed to 25% in 1998). Another indicator of public concern about crime relates to views about how the crime level has changed. Despite the decline in the crime rate, more than half of South Africans (53%) felt that crime has increased over the past three years in the areas where they live. These views were particularly prevalent among people in metropolitan and urban areas, and among Indian and white South Africans. In general, more people believed that property crime had increased (55%) than those who thought violent crime had gone up (47%). In terms of the ordering of crime types by their prevalence, little has changed since 1998. The most prevalent crimes five years ago were housebreaking, theft of livestock, theft of personal property, assault, fraud, theft out of motor vehicle and robbery. In 2003, the same crimes were among the top seven most prevalent offences, with the exception of fraud. In both years, property crimes occurred more frequently than violent crimes. A notable finding in the 2003 survey was that a significant proportion of South Africans (5.6%) reported being asked by a government official for a bribe in the form of money, a favour or a present in return for a service that the official was legally required to perform. This suggests that petty corruption was the second most prevalent crime type in the country. The bad news: less people feel safe Despite the decline in crime rates indicated by the victim surveys and the official crime statistics, South Africans feel less safe in 2003 than they did in 1998. Perceptions of safety can be measured in various ways, one of which is used internationally,2 and asks survey respondents how safe they feel when walking alone in their area during the day and after dark. In 2003, 85% of South Africans said they feel safe walking alone in their area during the day, while only 23% felt safe walking alone at night. On the positive side, the percentage feeling safe at night is higher than that recorded in other comparable site- based victim surveys in the country, including Cato Manor, Hillbrow/Inner Johannesburg, Cato Crest or Meadowlands.3 However, significantly less South Africans felt safe walking in their area at night than those surveyed in other developing countries. The ICVS found that on average, 60% of those surveyed in African countries, 56% in Latin American countries and 55% in Asian countries said they felt safe walking in their areas after dark. In South Africa only 23% said the same.4 Of more concern than the international comparisons, is that South Africans are much more fearful now than they were five years ago. During the day, the public felt generally as safe in 2003 as SA CRIME QUARTERLY No 8 JUNE 2004 19MISTRY 1998 Very unsafe Fairly safe Figure 1: Respondents’ feelings of safety when walking alone in their area during the day, 1998 and 2003 Source: ISS National Victims of Crime Survey, 2003 0 30 50 70 60 40 20 10 Very safe % 2003 25 60 60 25 10 9 5 6 A bit unsafe The survey results show that the risk of becoming a victim of property crime was greater than violent crime, and that people were more inclined to think property crime had increased than violent crime. Despite this, five of the top six crimes that South Africans were most afraid of, were violent, with murder topping the list even though it was among the least prevalent of the crimes investigated (Figure 3). It is possible that more sensational crimes such as murder and rape have a greater impact on perceptions, and are more intensely covered in the media. Moreover, how the police are believed to deal with particular crimes could contribute to public concerns about them. In general, these views indicate the types of crime that respondents thought they were most susceptible to, as well as their concerns about the impact of the offences. Although crime has levelled off since 1998, the results indicate that violence remains the key challenge as far as the public is concerned. Although fears about certain crimes did not match the risk of actually becoming a victim, public views SA CRIME QUARTERLY No 8 JUNE 200420 MISTRY about which crimes are most common were a closer match to reality. When asked “what one type of crime occurs most in your area?”, respondents were most likely to say housebreaking (38%), followed by robbery (14%), theft of property (10%), murder (7%), stock theft, bag snatching, and assault (all 6%), rape (4%), car theft (3%) and hijacking (2%). The order of crimes believed to be most common was fairly similar to the actual victimisation rates (Table 1), with the exception of murder and robbery, whose prevalence was overestimated by respondents. Impressions of police performance Respondents were asked about their physical access to the police, whether they had actually been to the nearest police station, and how they rated the performance of the police in their area. Access to the police was generally good: almost all South Africans (97%) knew where their nearest police station is, and two thirds were able to reach the police station within 30 minutes or less using their usual mode of transport.5 This should improve 1998 Very unsafe Fairly safe Figure 2: Respondents’ feelings of safety when walking alone in their area after dark, 1998 and 2003 Source: ISS National Victims of Crime Survey, 2003 Source: ISS National Victims of Crime Survey, 2003 0 30 50 70 60 40 20 10 Very safe % 2003 10 32 13 24 19 19 58 25 A bit unsafe 0 Not afraid 5 10 15 20 25 30 Livestock theft Pick pocketing/ bag snatching Theft of property Car hijacking Assault Robbery Sexual assault/rape House breaking Murder % Figure 3: The one type of crime that respondents were most afraid of in the area where they live 25 23 19 13 5 4 4 3 2 1 SA CRIME QUARTERLY No 8 JUNE 2004 21 doubt informs people’s opinions on the fear of crime and their reliance on the police for assistance. An important issue related to police visibility and performance is the level of reporting of crime by victims. High reporting rates reflect, among other things, levels of public confidence in the police. The reporting rate also gives an indication of how many crimes are never registered in the police’s official database. Generally, serious property and violent crimes are reported, while offences regarded as petty (such as pick-pocketing), that may cause embarrassment to victims when reporting (such as rape), or are believed to be a matter for the parties concerned and not the state (such as domestic violence) are often not reported to the police. This pattern of reporting was found to be true for the most prevalent crimes recorded in the 2003 survey, with the exception of robbery. The vast majority of car theft victims reported the crime to the police, no doubt for insurance purposes (Figure 4). Similarly, a smaller majority reported theft out of vehicles, and housebreaking. The reporting rate for the chances that victims will report crime to the police. Unsurprisingly, those living in the highly urbanised Gauteng and Western Cape provinces were closest to police stations, while those in rural Limpopo and Eastern Cape had to travel the furthest to reach their local station. Just under half (46%) of the respondents indicated that they had visited their nearest police station in the last three years. Given that those with first hand experience of dealing with the police are better placed to articulate their views on police performance, the opinions of these respondents are important. Of those who had been in contact with the police, more than half (56%) said it had changed their opinion of the police, and of these, 54% said their opinion had improved. A little more than one tenth (12%) claimed their opinion remained unchanged, while just over one third (35%) said it had made their opinion worse. Perceptions of police performance were also tested in a question to all respondents about how they think the police are doing in their area of residence. Just over half (52%) of South Africans said the police were doing a good job in their area, while more than two out of five (45%) thought they were doing a bad job. The main reasons cited for why the police are doing a good job were their commitment (25%), that they arrest criminals (24%), respond on time (23%) and come to the scene of a crime (15%). The main reasons for saying the police are doing a bad job included that they do not respond on time (35%), are corrupt (13%), don’t come into the respondent’s area (12%) and are lazy (11%). The importance of police response times in both sets of reasons indicates an area that is directly within the control of the police that could be worked on to improve public perceptions. Police visibility is also an important factor regarding perceptions of safety. When asked how often they see the police on duty and in uniform in their area, respondents were most likely (29%) to say they see a police officer at least once a day. Just more than a quarter (25%) said they see the police at least once a week. A major cause for concern is that one fifth (21%) reported that they “never” saw a police officer on duty in their area of residence. This no MISTRY Source: ISS National Victims of Crime Survey, 2003 0 Robbery 20 40 60 80 100 120 Livestock theft Damage to motor vehicles Theft Bicycle theft Assault Housebreaking Theft out of motor vehicle Car theft % of victims who reported Figure 4: Reporting to the police for crimes with rates >1% 97 64 57 55 44 41 36 36 29 10 assault is fairly high, considering that this is a crime that is often regarded as not important enough to bother reporting, or not a matter for the police to resolve. In the case of robbery however, the fact that only 29% of victims reported the offence is worrying, particularly considering that most of the robberies recorded in the survey were armed robberies and thus of a serious nature. Reporting rates have nevertheless improved since 1998 for several of the more prevalent crimes, again with the exception of robbery (Figure 5). The results suggest that confidence in the police as measured by reporting rates has grown in the past five years. It is however important to bear in mind that victims’ decisions about whether or not to report are based on a range of factors, some of which are not directly related to policing, such as the view that it is unnecessary to report, or the fear that the perpetrator will take revenge on the victim if he or she reports. Views of the courts Respondents were asked a similar set of questions about their physical access to the courts, as well as their views of court performance. As in the case of SA CRIME QUARTERLY No 8 JUNE 200422 MISTRY the police, access to the courts was generally good: more than two thirds (84%) of South Africans knew where the nearest magistrate’s court is located and just over half (51%) said they can get to the court within 30 minutes or less using their usual mode of transport. Access was better in the more urbanised provinces. Respondents living in the Eastern Cape and Limpopo were most likely to have to travel long distances, while those in Gauteng and Western Cape travelled for the shortest time. On the whole, slightly more South Africans (59%) felt the courts were performing their duties adequately than the 52% who said the police were doing a good job. Levels of satisfaction with the courts were even higher among those who had direct experience with the court system: of the one fifth (22%) who had been to court in the last three years, most (70%) were happy with the service provided by the state prosecutor/state advocate. A similar majority (71%) was happy with the magistrate or judge that presided over the case. All respondents, regardless of whether they had been to court or not, were asked whether they were satisfied with the way courts generally deal with perpetrators of crime. Just over half (51%) said they were, with almost as many (45%) expressing their dissatisfaction. The main reasons given for being satisfied were that the courts pass appropriate sentences (60%), have high conviction rates (22%) and are not corrupt (17%). Dissatisfaction centred on the courts being too lenient (34%), releasing perpetrators unconditionally (32%), not enough convictions (16%), and matters dragging on for too long (14%). These results indicate that sentencing was the main issue about which the public formed their opinions, both positive and negative, of the way courts deal with suspects. It is also revealing that the second most common reason for criticising court performance was that perpetrators are released “unconditionally”. This suggests that the public do not understand the bail and sentencing processes. Views about perpetrators of crime The above results suggest that the public favour stiff sentences for perpetrators. But who do South Source: ISS National Victims of Crime Survey, 2003 and Stats SA, 1998 0 20 7030 40 50 60 Robbery % of victims reporting to police Figure 5: Comparative reporting rates, 2003 and 1998, selected crimes 41 19982003 29 Theft of personal property 28 41 Assault 38 55 Housebreaking 57 57 Theft out of vehicle 57 64 Africans think is responsible for committing most crime, and what are the motivations of these criminals believed to be? Public opinion on the issue is likely to be informed, considering the high number of people who know someone in their area who makes a living from crime: 29% of respondents admitted to this, which is not that surprising given that crime rates are relatively high. Respondents were further asked about the residency and origin of the perpetrators. The responses clearly indicate that contrary to popular opinion, the vast majority of South Africans believe that people born in South Africa are responsible for most crime. Only 4% thought that most crime was committed by foreigners. Respondents were also of the view that most violent and property crime is carried out by people who live in their area, rather than by ‘outsiders’. When asked about perpetrators’ motivation for committing crime, the most frequent answers for both property and violent crimes were “greed” and “non-financial motives” as opposed to “real need”. Real need was however almost as common an explanation for property crime as the other reasons (Figure 6). Although a common perception is that crime is caused by poverty, these results suggest that the public think otherwise. No matter what the motives for crime were believed to be, most South Africans said developmental solutions are most important for solving the problem. When asked which one of three options (crime prevention and law enforcement including more police; the judiciary and courts including harsher sentences, punishment and prisons; and social development including job creation) government should spend money on to reduce crime in their area, most South Africans opted for social development. A further one quarter said money should be spent on crime prevention and law enforcement, with the remainder identifying the judiciary and courts as important (Figure 7). Conclusion The results of South Africa’s second national victim survey, as well as the police statistics, show that crime rates have either decreased or levelled off over the last five years. However, according to the SA CRIME QUARTERLY No 8 JUNE 2004 23 Violent Figure 6: Views on what motivates most perpetrators of property and violent crime Source: ISS National Victims of Crime Survey, 2003 Source: ISS National Victims of Crime Survey, 2003 0 30 40 20 10 25 45 35 15 5 Greed % Property 36 39 Non-financial motives 34 43 Real need 30 18 Violent crime Figure 7: ‘Which one of the following should government spend money on to make your area safe from crime?’ 0 10 20 30 40 50 60 70 Social development (create employment) % Property crime 63 50 Crime prevention and law enforcement (more police etc) 23 24 The courts and prisons (punishment, harsher penalties) 15 26 MISTRY victim survey the public’s fear of crime has simultaneously increased. This counter intuitive trend may be explained by a number of factors such as increasing public awareness of other people’s victimisation and the high level of violence that typifies some criminality. However, more research is required in order to understand the complex dynamic between the increasing fear of crime and decreasing crime rates.6 Endnotes 1 The full report of the survey results will be published shortly. See P Burton, T Leggett, A Louw, D Mistry and H van Vuuren, National Victims of Crime Survey: South Africa 2003, ISS Monograph Series, July 2004, Institute for Security Studies, Pretoria, forthcoming. 2 This question is asked in the International Crime Victim Surveys (ICVS) conducted by the United Nations Interregional Crime and Justice Research Institute (UNICRI) over the past 15 years in 24 industrialised countries and 46 ‘countries in transition’. 3 See T Leggett, Rainbow Tenement: Crime and policing in inner Johannesburg, ISS Monograph Series, No 71, ISS, Pretoria, April 2003. 4 The ICVS results are reported in A A del Frate and van Kesteren, The ICVS in the developing world, International Journal of Comparative Criminology, 2(1), de Sitter Publications, 2003, pp 57-76. 5 Since the 1997/1998 financial year 28 police stations, 13 satellite stations and 9 contact points have been established (figures obtained from SAPS Efficiency Services, March 2004). 6 A series of focus group discussions will be undertaken in due course by the ISS. SA CRIME QUARTERLY No 8 JUNE 200424 MISTRY