CQ No. 25 SA Crime Quarterly no 34 • December 2010 3 The state of crime in South Africa Johan Burger, Chandré Gould and Gareth Newham* jburger@issafrica.org, cgould@issafrica.org, gnewham@issafrica.org This article presents an analysis of the 2009/10 SAPS cr ime statistics, which are released by the South African Police Service in September each year. The statistics for the 2009/10 period show an overall increase in crime at a national level that is driven by increases in five categories of crime: shoplifting, commercial crime, residential and business burglaries, and theft from motor vehicles. While the statistics suggest that violent crime has decreased, there are a number of questions about the accuracy of the statistics. The article discusses this issue and reflects on how the accuracy and reporting of crime statistics could be improved. Since 1996 the South African Police Service (SAPS) has released the crime statistics for South Africa annually, in September.1 The statistics are disaggregated both by crime type and by location (i.e. nationally, provincially and at a precinct level). The statistics are usually six months out of date upon their release, and doubts about their veracity tend to bedevil the annual announcement of the crime figures. Yet, in the absence of regular victimisation survey data, the SAPS statistics are the only annual and relatively comprehensive measure of the change in crime trends over time. These figures also contribute towards influencing South Africans' perceptions of their level of safety. For this reason it is worthwhile to critically assess what the statistics have revealed over time and the factors that affect their accuracy. OVERALL CRIME LEVELS2 A longitudinal analysis of total national crime recorded annually shows that overall crime levels peaked in 2002/03, after which there was a gradual decline until 2007/08 when crime started to rise once again (see Figure 1). There are no clear explanations as to why total crime decreased by 25 per cent until 2007/08 before increasing by four per cent since then. In this regard South Africa is not exceptional. Efforts to explain the dramatic reduction in violent crime rates in the United States in the ten years between 1999 and 2010 have been fraught with controversy, and analysts have arrived at vastly different explanations for the drop.3 What the statistics reveal however, is that 97 per cent of the increase in total crime levels over the past two years is being driven by increases in the following five property-related crime categories: • Shoplifting increased by 32 per cent or by 21 642 cases; * Johan Burger and Chandré Gould are senior researchers, Crime and Justice Programme, Institute for Security Studies. Gareth Newham is the programme head, Crime and Justice Programme, Institute for Security Studies. An analysis of the SAPS crime statistics for 2009/10 CQ No. 34 - Dec 2010 12/17/10 9:31 AM Page 7 between 2003/04 and 2007/08. Burglaries at businesses began increasing a year earlier, during the 2005/06 period, but unlike residential burglary, there was no history of a sustained decrease in this type of crime. Yet, if the economic downturn were behind increasing crime levels, it can be expected that this upward trend in property crime would be 4 Institute for Security Studies Figure 1: Total national crime levels 1994/95–2009/10 3 000 000 2 500 000 2 000 000 1 500 000 1 000 000 500 000 0 94 /9 5 95 /9 6 96 /9 7 97 /9 8 98 /9 9 99 /0 0 00 /0 1 01 /0 2 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 Violent interpersonal crime Robbery Property crime Other Figure 2: Shoplifting and commercial crime 2003/04–2009/10 100 000 80 000 60 000 40 000 20 000 0 03/04 04/05 05/06 06/07 07/08 08/09 09/10 Commercial crime Shoplifting • Commercial crime increased by 30 per cent or by 19 556 cases. (Since 2004/05 the absolute numbers of this type of this crime have increased by 57 per cent); • Residential burglary increased by eight per cent or by 18 724 cases; • Theft out-of-motor vehicles increased by eight per cent or 9201 cases; and • Non-residential burglary (often referred to as business burglary) increased by 14 per cent or 8 778 cases (see Figure 2). Over this two-year period, murder decreased by 1 653 cases (8,9 per cent), attempted murder decreased by 1 385 cases (7,4 per cent), assault with intent to inflict serious bodily harm reduced by 4 811 cases (2,3 per cent) and aggravated robbery reduced by 4 557 cases (3,9 per cent). Since property crime has shown an increase while interpersonal violent crime has decreased, it may be tempting to blame the impact of the global economic downturn of the last three years for the increase in overall levels of crime in South Africa. Residential burglary began to increase in 2007/08, after having declined over the four-year period CQ No. 34 - Dec 2010 12/17/10 9:31 AM Page 8 mirrored in other countries. But this is not the case. The United States and Britain, for example, have seen an overall decrease in crime rates, in particular in relation to property crimes, in spite of the economic recession.4 According to the British Crime Survey report these results were contrary to the expectation that property crime would increase during times of recession.5 A number of speculative reasons were offered for these unexpected results, for example that the downturn may have been the consequence of target hardening through measures to increase security at homes and of vehicles. This was backed by survey data that showed that in England and Wales households with 'less than basic' security measures were six times more likely to have been victims of burglary than those with 'basic' security measures and ten times more likely than homes with 'enhanced' home security measures.6 The British Home Office report cites other studies and hypotheses for the change in crime rates, but concludes that while there is 'broad' support for the 'improved security measures' theory, there is no scientifically acceptable explanation for the reduction in property crime during a recession.7 Violent ‘social fabric’ crimes So-called 'social fabric' crimes include murder, assault, sexual offences and domestic violence. The police refer to these crimes as 'social fabric' crimes because they tend to take place in private spaces between people who know each other and, as such, are believed to be caused by social and interpersonal factors. As a result they don't respond to traditional forms of visible policing such as patrols, roadblocks or 'crackdown' operations.8 The SAPS statistics suggest that over the past nine years these types of crimes have either reduced or stabilised, with the exception of attempted murder that rose during the 2002/03 reporting period (Figure 5). However, serious doubts remain about the veracity of the assault, attempted murder9 and even the murder statistics.10 It is therefore difficult to offer a clear reason for the changes in trends of these categories. The SAPS statistics show that since 1994/95, murder has reduced by 50 per cent. The decrease of 7,2 per cent in the absolute number of murders in the past year is the third largest year-on-year decline since 1995. One of the factors contributing to the decline in murders may be the 6,3 per cent decrease in the number of aggravated robberies. Since almost 16 per cent of murders in South Africa occur as a result of robbery, the decrease in aggravated robberies may have contributed to the decline in murders. Yet, the decline in murders could also be a consequence of murders having been recorded as inquests (in other words as deaths requiring post- mortem examination in order to determine the cause of death). In a written reply to a parliamentary question, the Minister of Police11 provided data showing that the decrease in murders over the last three years has been accompanied by an increase in inquests (see Table 1). It has subsequently been reported that a special SAPS task team was instructed to urgently probe whether police stations have incorrectly registered murder cases as inquests.12 It is notable that over the past 15 years, attempted murder has shown curious ups and downs. Attempted murder substantially increased between 2000/01 and 2003/04, then dramatically SA Crime Quarterly no 34 • December 2010 5 Burglary at non- residential premises Burglary at residential premises Figure 3: Residential and business burglaries 2003/04–2009/10 03/04 04/05 05/06 06/07 07/08 08/09 09/10 350 000 300 000 250 000 200 000 150 000 50 000 0 CQ No. 34 - Dec 2010 12/17/10 9:31 AM Page 9 6 Institute for Security Studies decreased by eight times the decrease in the murder rate. It is possible that this unusual trend can be explained by police recording practices instead of any fundamental (but temporary) shift in the crime of attempted murder during that period. Another peculiarity can be found in the relationship between the trend in murder and the trends for common assault and assault with intent to cause grievous bodily harm (Figure 6). Because most murders start out as assaults it can reasonably be assumed that there should be some correlation between the two categories. Although incidents of assault decreased by well over 20 per cent between 2002/03 and 2009/10, the figures have now stabilised, showing slight increases over the past year, during which murders decreased substantially. According to Bruce,13 the assault statistics are the most likely to be unreliable as there is likely to have been a significant under- reporting of assault by police seeking to improve their performance ratings. He argues that assault is a crime that would be easy for the police to under-report.14 Aggravated robbery Aggravated robbery comprises seven sub- categories (street robbery, car hijacking, truck hijacking, cash-in-transit robbery, bank robbery, house robbery and business robbery) and remains a significant crime threat to all sectors in South Africa. This is because robbery has a major impact on perceptions of safety due to its violent nature, and also because it has a significant economic impact. The economic impact of aggravated robbery is both direct (stolen cash and property) and indirect (the cost of security, insurance and limitations on business investments).15 Robbery is the crime type most likely to cause a decline in feelings of safety as measured by three national victims of crime surveys since 1998.16 For example, whereas 56 per cent of respondents in 1998 indicated that they 'felt safe' walking in their own areas after dark, this figure dropped to only 21 per cent in 2007.17 This correlated with the 2007/8 18 487 29 464 2008/9 18 148 -1.8 30 558 3.7 2009/10 16 834 -7.2 31 265 2.3 Murder % change Inquests % change Table 1: Recorded murders versus inquests 2007/08–2009/10 Figure 5: Murder and attempted murder: 1994/95–2009/10 40 000 35 000 30 000 25 000 20 000 15 000 10 000 5 000 0 94 /9 5 95 /9 6 96 /9 7 97 /9 8 98 /9 9 99 /0 0 00 /0 1 01 /0 2 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 Murder Attempted murder 26 801 26 563 28 159 29 553 28 198 28 108 31 293 35 861 30 076 20 142 17 410 18 795 26 883 24 516 20 553 18 298 25 965 26 877 25 470 24 486 25 127 21 758 21 405 21 553 19 824 18 793 18 54519 202 18 487 18 148 16 834 22 604 CQ No. 34 - Dec 2010 12/17/10 9:31 AM Page 10 SA Crime Quarterly no 34 • December 2010 7 2007 victimisation survey confirming that robbery rates had increased while victimisation rates for other crimes had decreased. Although total aggravated robbery decreased overall by 6,3 per cent over the 12 month period 2008/09–2009/10, it has remained at a consistently high level for most of the past decade (see Figure 7). Interestingly, when the National Crime Prevention Strategy (NCPS) was launched in 1996 with the intention of drastically reducing crime, aggravated robbery was at its lowest recorded rate.18 Yet, the post-1996 trends, in robbery particularly and in crime in general, demonstrate that this strategy was not properly implemented. The total aggravated robbery trend conceals the trends in the sub-categories of robbery that are of particular concern to many South Africans. This is particularly the case with the trends in so- called 'trio crimes': house robbery, business robbery and car hijackings (see Figure 8). The 2009/10 statistics suggest that levels of these crimes may have stabilised after a substantial year-on-year increase since 2003/04. In the 2009/10 period vehicle hijacking decreased by 6,8 per cent at a national level and there was also a significant national decrease in vehicle theft (18,5 per cent over the last seven years). This may indicate that technology to reduce vehicle theft and hijacking has improved along with the ability of the police to address vehicle theft syndicates. House robbery, that increased by 100 per cent since 2003/04, can also be considered to have stabilised for the first time in five years with a small (1,9 per cent) increase in 2009/10. Business robbery increased by 295,3 per cent between 2003/04 and 2009/10. The rate of Figure 6: Trends for assault GBH and assault common 1994/95–2009/10 300 000 250 000 200 000 150 000 100 000 50 000 0 94/95 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 06/07 07/08 08/09 09/10 Figure 7: Aggravated robbery 1994/95–2009/10 160 000 140 000 120 000 100 000 80 000 60 000 40 000 20 000 0 Assault common Assault grievous bodily harm 94 /9 5 95 /9 6 96 /9 7 97 /9 8 98 /9 9 99 /0 0 00 /0 1 01 /0 2 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 215 671 223 097 231 497 234 819 237 818 261 804 275 289 264 012 266 321 260 082 249 369 226 942 218 030 210 104 203 777 205 293 200 248 206 006203 023 201 317 203 678 232 024 248 862 261 886 282 526 280 942 267 857 227 553 210 057 198 049 192 838 197 284 CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 11 8 Institute for Security Studies increase has however slowed over the past year, with an increase of only 4.4 per cent. The change in the trends of these crimes may be attributable to improved policing, particularly improved police intelligence and investigations targeting the perpetrators of these crimes. Provincial trends in the 'trio crimes' From a crime analysis point of view, it is instructive to consider the picture that emerges from the provincial crime rates. Gauteng, which records half of the national total of trio crimes, recorded a 1,7 per cent reduction in house robberies, a 2,8 per cent reduction in car hijacking and a small increase of 2,6 per cent in business robberies over the past 12 months. KwaZulu-Natal, which records the second highest number of trio crimes of all the provinces, experienced a decrease of 17,3 per cent in business robberies, a 0,5 per cent decrease in house robberies and an 8,5 per cent decrease in hijackings. The picture is notably different in the other provinces where residential and business robberies have increased at a rate higher than the national average. In the Eastern Cape business robberies jumped by 49,6 per cent (851 to 1 273 cases) and the Northern Cape recorded an increase of 21,5 per cent (121 to 147 cases). Although car hijacking decreased nationally by 6,8 per cent there was a marked increase in the Free State from 255 cases to 316 cases (an increase of 23,9 per cent), and the Northern Cape, from 7 to 13 cases (85,7 per cent). Limpopo, however, showed the greatest increases in trio crimes, despite having had fairly low levels of these types of crimes in the previous six years (Figure 9). During 2009/10 Limpopo was the only province where increases were recorded in murder (1,5 per cent), street robbery (2 per cent), house robbery (13,6 per cent) and business robbery (19,1 per cent). This strongly suggests that there was a problem with policing in Limpopo since, with the exception of murder, these are the types of crimes that the police should be able to have the greatest impact on. Police impact on crime The South African government's concern about the crime challenge is reflected in the substantial increases in the budget of the SAPS and numbers of police officials that have been recruited. Since 2002/03 more than 60 000 police personnel have Figure 8: Trio crimes 2002/03–2009/10 (total numbers) 20 000 18 000 16 000 14 000 12 000 10 000 8 000 6 000 4 000 2 000 0 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 9 603 9 351 9 391 10 173 12 761 14 481 18 438 18 706 14 691 13 793 12 434 12 825 13 599 14 201 14 915 13 902 5 498 Car hijackings Business robbery House robbery 3 677 3 320 4 287 6 689 9 862 13 920 14 534 CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 12 SA Crime Quarterly no 34 • December 2010 9 been recruited, representing an increase of 44,4 per cent (Figure 10). According to the Minister's budget speech in Parliament in February 2008 the target was to increase SAPS numbers to over 200 000 in 2010/11.19 In addition, in the six-year period between 2004 and 2010, the police budget increased by almost 132 per cent from R22,7 billion to R52,6 billion (Figure 11). A large proportion of this budget went to recruiting additional personnel. Large investments were also made in technology and training. Yet, organised violent crime that is most susceptible to effective policing strategies remains unacceptably high. Fortunately, it has recently been recognised that swelling the ranks of the police alone is not sufficient to reduce violent crime. The National Commissioner of Police, General Bheki Cele, while addressing Parliament's Portfolio Committee on Police on 14 September 2010, acknowledged that the 'rush for new recruits' and the 'common perception' that the police were the 'last resort for those who failed at everything else', had lowered standards in the police.20 It would be reasonable to expect that the increase in the SAPS budget and personnel numbers, and their concomitant gains in vehicles and other equipment should have a corresponding impact Figure 9: Violent crime trends in Limpopo 2003/04–2009/10 (rates per 100 000) 70 60 50 40 30 20 10 0 03/04 04/05 05/06 06/07 07/08 08/09 09/10 03/04 04/05 05/06 06/07 07/08 08/09 09/10 Aggravated robbery 20 15 5 0 Business robbery MurderHouse robbery Figure 10: Police personnel expansion 2002/03–2010/11 Figure 11: Police budget growth 2002/03–2010/11 250 000 200 000 150 000 100 000 50 000 0 134 857 201 000 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 10 /1 1 Police numbers 60 50 40 30 20 10 0 22,7 52,6 02 /0 3 03 /0 4 04 /0 5 05 /0 6 06 /0 7 07 /0 8 08 /0 9 09 /1 0 10 /1 1 Police budget in SA Rand CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 13 10 Institute for Security Studies on those types of crimes that have been shown to respond to improved policing, such as the trio crimes. Yet, so far this has not been the case. The police did achieve a level of success in terms of visible policing, high density policing, effective targeted investigations and reassuring the public during the recent FIFA World Cup in South Africa.21 This has raised questions in South Africa why the same quality of policing is not evident at other times, even bearing in mind that the World Cup was clearly an exceptional period within which the police could focus attention on visible policing around stadia. However, there are many reasons not to place the blame for sustained high crime levels on the police alone. In this regard it is worth considering the results of police case docket analyses (keeping in mind that the results reflect primarily on criminal cases where the perpetrators were either caught or at least known). Indeed, it is to this analysis that the police themselves have turned to explain the problem. SAPS case docket analyses have shown that the victims and perpetrators were known to each other in between 70–80 per cent of murders, 60 per cent of attempted murders, 75 per cent of rapes and 90 per cent of assaults.22 Approximately 65 per cent of murders were associated with 'social behaviour' such as drug and alcohol abuse and 'only' 16 per cent happened as a consequence of other crime, such as robbery.23 Policing can have little, or no effect on these kinds of crimes. This is supported by international research that argues that so-called social fabric crimes are driven by social conditions and occur within social settings outside the control of the police. For example, Newburn and Reiner have made the following observation: 'Crime is the product of deeper social forces, largely beyond the ambit of any policing tactics, and the clear-up rate is a function of crime levels and other aspects of workload rather than police efficiency.'24 The challenge for the police over the next few years will be enormous if they are expected to justify the increased expenditure of public money. Yet it is still far from clear that high expectations or improved policing will, or even can, have the desired effect. CAN WE RELY ON THE SAPS STATISTICS? Each year the release of the crime statistics by the Minister of Police leads to media hype about the state of crime in South Africa. Yet the extent to which the official crime statistics allow for an accurate assessment of the state of crime in South Africa remains contentious. David Bruce from the Centre for the Study of Violence and Reconciliation (CSVR) has argued that the implication of suspected widespread under-reporting of crime by the police, supported by proven individual cases, is that 'current crime statistics cannot be regarded as a reliable indicator of trends in crime, particularly in violent crime.'25 But since there is also no way to reliably ascertain the extent to which crimes have been manipulated or to the extent to which under- reporting affects the reliability of violent crime trends, these statistics are the only official indicators of trends in crime in South Africa.26 Reliable crime statistics are important, not least because of the role they may play in informing the perceptions of safety of citizens. If citizens have reason to suspect the validity of the crime statistics, they may be more fearful of crime than if they could trust the statistics, because they will believe that the police are hiding the real extent of the problem. This in turn may have a negative effect on civilian trust in the police. In addition, inaccurate statistics pose a problem for crime analysts and the police themselves, who need accurate information upon which to base their crime reduction strategies. Addressing this problem requires several interventions. An independent audit of South Africa's crime recording systems and crime statistics would be one way in which to restore public trust in the crime statistics and determine the reliability of the figures. Another way to address the shortcomings of the crime recording system would be to conduct a regular crime survey (preferably annually) that is capable of independently measuring the level of CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 14 SA Crime Quarterly no 34 • December 2010 11 crime, irrespective of whether crimes are reported to the police or not. A good example of such an independent instrument is the annual British Crime Survey (BCS) that supplements the police recorded crime statistics and thus provides a basis upon which to assess the accuracy of the police's figures. The British Crime Survey is conducted annually amongst a representative sample of 45 000 households in England and Wales.27 According to the July 2010 British Home Office report on crime in Britain, the British Crime Survey is a better measure of trends in crime over time for those crime types that are included in the survey, because 'it is unaffected by changes in levels of public reporting or in police practice in recording crime.'28 In 2006, an independent review group was appointed to advise the UK's Home Secretary '... on what changes could be made to the production and release of crime statistics...'.29 The Review Group summarised among their main conclusions the following: The focus must shift from the publication... of the aggregate national picture to a system of communication which encompasses local data at local level... [and] [g]overnance, management and organisation of the police and Home Office environments in which crime statistics are produced and reported must be revised to provide the public with complete assurance of actual and perceived independence and integrity of the statistics.30 The Home Office now releases crime statistics four times a year, based both on police recorded crime and the results of the BCS.31 There have been three national victims of crime surveys in South Africa, with the last two being conducted by the Institute for Security Studies in 2003 and 2007. Fortunately, government has recognised the importance of these surveys and Statistics SA will be conducting one in 2011 with a large sample of 30 000 households. It is important for citizens to have an accurate picture of the extent of crime, both nationally and in the areas they live. This would include knowing the extent to which crime categories are increasing or decreasing in as close to real time as possible (e.g. weekly or monthly). This information should be made available more regularly than once a year and is important for a variety of reasons: • Knowledge about the extent and nature of crime enables citizens to make informed decisions about, for example, where they invest, where they buy or rent property, and what they do to improve their personal or property security; • If citizens are properly informed and understand exactly what crimes affect their areas they may be encouraged to become involved in local crime prevention programmes that target these types of crime specifically; • Providing regular, reliable information about the extent and nature of crime is one way that the state can improve public trust in the police; • Also, localised data, together with localised responses by the police acting with residents and community-based organisations such as CPFs, are much more likely to result in a reduction of crime in those neighbourhoods. Such reductions and successes are crucial to improving public confidence and trust in the police. CONCLUSION The 2009/10 statistics strongly indicate that it is not more policing that is needed, but smarter and more accountable policing, if we are to see a positive change in the rate of those types of crime that do respond to policing. The Minister of Police's statement on 20 October 2010 that he intends to improve partnerships between the police, business, civil society and communities is very welcome.32 However, effective partnerships require trust and sharing of information. It is therefore important that the credibility of the SAPS crime statistics is addressed and that up to date crime statistics are made available to the public regularly so that CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 15 communities can make effective use of the information and work with the police in a meaningful and goal-directed manner. To comment on this article visit http://www.issafrica.org/sacq.php NOTES 1 There was one exception to this. In 2008 the Minister of Safety and Security, Charles Nqakula, allowed the statistics to be released twice – in June and in September. 2 All the crime figures used in this article are from the Annual Report of the South African Police Service 2009/10. SAPS, Annual Report of the South African Police Service 2009/10, Pretoria, 2010, pp.1–35. Available at www.saps.gov.za (accessed 10 November 2010). Crime statistics older than 2003/4 were taken from the SAPS Annual Report 2003/04, p.19. 3 See Steven D Levitt, Understanding Why Crime Fell in the 1990s: Four Factors that Explain the Decline and Six that Do Not, Journal of Economic Perspectives, 18 (1) 2004, 163–190. Also see FE Zimring, The Great American Crime Decline, New York, Oxford University Press, 2007. 4 Johan Burger and Chandré Gould, Crime Stats: Not all bad news – but little cause to celebrate: A mixed bag of data provides some encouraging trends – but the need for quality police continues, Pretoria News (Focus), September 10, 2010, 13; and Gareth Newham, Johan Burger and Chandré Gould, Stats useless unless we know the 'why', Pretoria News (Focus), September 10, 2010, 13; Home Office Statistical Bulletin, 2–3. 5 John Flatley, Chris Kershaw, Kevin Smith, Rupert Chaplin and Debbie Moon, Crime in England and Wales 2009/10: Findings from the British Crime Survey and police recorded crime, Home Office Statistical Bulletin, July 2010, available at rds.homeoffice.gov.uk/rds/pdfs10/hosb1210.pdf (accessed 19 October 2010), 2. 6 Ibid. 7 Ibid, 3. 8 See for example SAPS, Annual Report of the South African Police Service 2009/10, Pretoria, September 2010, 10. 9 David Bruce, 'The ones in the pile were the ones going down': the reliability of current violent crime statistics, South African Crime Quarterly 31, March 2010, 9–19. 10 Shanaaz Eggington, Police probe crime figures, Sunday Times, 31 October 2010. 11 Minister of Police, Written reply to parliamentary question number 2716, reply submitted 12 October 2010, Mrs DA Schafer (DA) to the Minister of Police. Available at http://www.politicsweb.co.za/politicsweb/ view/politicsweb/en/page72308?oid=204385&sn=Mark etingweb+detail&pid=90389 (accessed 2 November 2010). 12 Eggington, Police probe crime figures. 13 Bruce, The ones in the pile. 14 Ibid. 15 Small Business Partnership. The Impact of Crime on Small Businesses in South Africa. A Study Commissioned by the South African Presidency, July 2008. 16 R Pharoah, National Victims of Crime Survey: Overview of key findings, ISS Paper 175, December 2008, Institute for Security Studies, Pretoria. 17 Ibid, 7. 18 National Crime Prevention Strategy, Department of Safety and Security, South Africa, May 1996, www.info.gov.za/otherdocs/1996/crime1.htm (accessed 20 October 2010). 19 T Manual, Budget vote speech by the Minister of Finance, Parliament of South Africa, 20 February 2008, Cape Town, http://www.treasury.gov.za/documents/ national budget/2008/default.aspx (accessed 29 October 2010). 20 Police sacrifice quality for quantity – Cele, News24, 14 September 2010. 21 J Burger, SA police kept its promise, the African.org, alternate-monthly magazine of the Institute for Security Studies, Issue 9, 2010, Pretoria, 50–51. 22 SAPS Annual Report, 7. 23 Ibid. 24 T Newburn & R Reiner, Policing and the Police, in M MaGuire, R Morgan & R Reiner (Eds), The Oxford Handbook of Criminology (Fourth Edition), Oxford: University Press, 910–945 (943). 25 Bruce, The ones in the pile, 9–17. 26 J Burger, Worrying trends: The official 2008/09 South African crime statistics, SA Crime Quarterly 30, December 2009, 3–11. 27 Flatley et al, Crime in England and Wales. 28 Ibid. 29 Crime Statistics: An independent review, Carried out for the Secretary of State for the Home Department, United Kingdom, November 2006, rds.homeoffice.gov.uk/.../crime-statistics-independent- review-06.pdf (accessed 19 October 2010). 30 Ibid, iii. 31 Home Office Statistical Bulletin, July 2010. 32 N Mthethwa, Remarks by the Minister of Police at a conference of the Consumer Goods Council of South Africa, titled 'Building Success Together in Changing Times', 20 October 2010, Sandton, http://www.polity. org.za/art/sa-mthethwa-remarks-by-the-minister-of- police-at-the-consumer-goods-council-of-south-africa- conference (accessed 29 October 2010). 12 Institute for Security Studies CQ No. 34 - Dec 2010 12/17/10 9:32 AM Page 16