SA CRIME QUARTERLY No 18 DECEMBER 2006 1
THE START OF A
‘CRIME WAVE’?
The 2005/06 official
crime statistics in
context
Despite growing fears of a new ‘crime wave’, the public have access to official crime statistics only up to
March 2006. Although much of the news up to this point was positive, perceptions of the statistics and of
government’s position on crime in the last six months have been characterised by scepticism and alarm. The
Minister of Safety and Security assured us that next year the statistics will be released soon after the end of
the financial year. This, together with a much clearer strategy and commitment from government as a whole
for dealing with crime, may help to assuage public anxiety.
T
he release of the official crime statistics by
the South African Police Service (SAPS) for
the last financial year (April 2005–March
2006) on 27 September this year was bound to be
controversial. The past six months have seen public
fears growing about the development of a new
crime wave. Widespread media coverage
particularly of violent robberies (some of which
resulted in rape and murder) in Western Cape and
Gauteng, together with the infamous Jeppestown
(Johannesburg) shootout between police and about
24 gangsters in June in which 12 people died
(including four police officers killed execution-style),
have fuelled concerns about an upsurge in crime.
These concerns were confirmed by the Minister of
Safety and Security who stated soon after the
Jeppestown incident1 and again at the press
conference at which the statistics were released, that
some crimes have indeed increased in the past six
months. Together with the uncharacteristically
candid revelations from leading members of the
short-term insurance industry about a 30-40% surge
in claims for car hijackings between March and
June 2006 compared to the same period last year,
there is now little doubt that in some parts of the
country, crimes like robbery have increased
markedly this year.
Bearing in mind that the latest release of official
data only covered the 12-month period up to
March 2006, and the limitations inherent in official
recorded crime data,2 what did the statistics show?
The overall crime picture
In brief, total crime levels continued their
downward trend of the past three years, with
several serious crimes following suit. The trends
that are of concern relate to crimes that are usually
highly organised, such as cash-in-transit robbery,
car hijacking and car theft.
Figure 1 below shows the total number of crimes
recorded by the SAPS between 1994/95 and
2005/06. Over this 12-year period, the total for the
21 most serious crimes in South Africa increased by
7%, from 2,022,899 to 2,168,952 offences. Crime
levels peaked in 2002/03 when 2,629,137 crimes
Antoinette Louw, Institute for Security Studies
alouw@issafrica.org
Figure 1: Total crime in SA (21 most serious crimes)
SA CRIME QUARTERLY No 18 DECEMBER 20062 LOUW
murder (which has declined steadily over the past
12 years).
Burglary
Between 1994/95 and 2005/06, the rate of burglary
of residential premises (measured per 100,000 of
were recorded. Since then, however, overall crime
levels have dropped steadily (by an average of 6%
per year), and between 2002/03 and 2005/06 crime
in the country decreased by 18%.
The majority of offences recorded in South Africa are
property crimes. Over a third (35%) are violent (9%
of which are robberies, and the remaining 26%
interpersonal violence like murder, rape and assault)
(Figure 2). The 16% of crimes categorised as ‘other’
in Figures 1 and 2 constitute arson, malicious
damage to property, the illegal possession of firearms,
drug related crimes, driving under the influence of
alcohol, and commercial crime. Over the past 12
years, the proportion of property crime has decreased
slightly while that of robbery has increased slightly.
The proportion of violent crime recorded in this
country is high, and there can be little doubt that the
nature and extent of violence is what sets South
Africa apart from other countries.
The overall crime picture sketched above presents
some good news considering the steady decrease
since 2002/03. This also holds true for several types
of serious crime, notably burglary, robbery and
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
0
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
Other
Property
Robbery
Violent
interpersonal
1994/95
27,993
110,807
11,744
51,745
1995/96
27,494
112,116
12,285
53,779
1996/97
27,970
108,278
11,683
54,522
1997/98
28,428
112,777
12,798
54,564
1998/99
28,186
121,343
15,760
55,081
1999/00
29,686
128,527
17,352
60,360
2000/01
30,019
136,400
20,393
63,356
2001/02
30,559
137,262
20,694
64,057
2002/03
31,428
141,891
22,844
66,750
2003/04
32,733
134,366
22,920
65,295
2004/05
34,232
120,006
21,761
62,577
2005/06
34,836
106,781
19,444
55,832
Figure 2: Categories of crime in 2005/06
Other
16% Violent
interpersonal
26%
Source: Annual Report of the SAPS for
2005/2006, Pretoria, 27 Sept 2006.
Property
49%
Robbery
9%
N
u
m
b
er
SA CRIME QUARTERLY No 18 DECEMBER 2006 3LOUW
The former seems plausible given that businesses are
likely to insure their property which in turn inclines
them to report burglaries to the police. However,
without more data on the size and nature of the
businesses concerned, it is difficult to assume that
most of them would have had insurance.
Robbery overall
The rate of robbery overall (both ‘aggravated’ and
‘common’) is significantly higher now than 12 years
ago. The rates for common robbery and aggravated
robbery increased by 89% and 16% respectively
between 1994/95 and 2005/06. But in both cases,
rates peaked around 2003, and have since dropped
fairly steadily. Aggravated robbery for example,
declined from 288 per 100,000 in 2003/04 to 255 in
the population) decreased by 6% in the country.
Like many other major crime types, the trend over
the 12-year period shows that residential burglaries
increased up to 2002/03, after which a steady
decline has been recorded. In 2002/03 the rate of
burglary was 704 per 100,000; by 2005/06 this had
dropped to 559 (Figure 3).
The trend for burglaries of business premises is
similar, although these burglaries peaked earlier than
residential burglary (in 1998/99), and have since
decreased dramatically (Figure 4). Significantly, the
number of business burglaries recorded by police in
2005/06 was 38% less than that recorded in
1994/95. This represents either a significant
improvement in the protection of business premises,
or substantial under-reporting by the sector.
Figure 3: Residential burglary rate
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
550
600
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
650
700
750
596
629
603 611
653
673
694
675
704
645
593
559
N
o
p
er
1
0
0
,0
0
0
p
eo
p
le
Figure 4: Business burglary
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
87,600
90,294
N
u
m
b
er
0
50,000
60,000
70,000
80,000
100,000
90,000 87,377
87,153
94,273 93,077
91,445
87,114
73,975
64,629
56,048
54,367
SA CRIME QUARTERLY No 18 DECEMBER 20064 LOUW
2005/06 (Figure 5), while common robbery
decreased from 223 in 2002/03 to 159 per 100,000
in 2005/06.
While the declines in the robbery rate in recent
years are encouraging, robbery remains one of the
most problematic crimes in terms of both the trends
and the problems it poses for analysis.
As far as analysis goes, the 2003 National Victims of
Crime survey conducted by the ISS found that
robbery was least likely to be reported to the police
of the nine major crime types covered in the survey.3
Only 29% of victims reported robbery – something
which cannot be attributed to their being ‘less
serious’ robberies since detailed analysis showed
that most of the robberies discussed in the survey
were of a serious nature (ie. they were committed
with a weapon). Complicating the analysis further is
that while a large chunk of robberies are believed to
be of cell phones,4 neither the police nor the
industry have provided clear figures to verify the
claim and its impact on the trends.
Aside from the problems with the robbery data,
there can be no denying that robbery is one of South
Africa’s most serious crime challenges. Aggravated
robbery in particular, which includes incidents
ranging from muggings at knifepoint, to car
hijackings, home robberies, bank robberies,
robberies at shopping malls and cash-in-transit (CIT)
heists, create widespread fear for individuals,
tourists, local businesses and foreign investors.
These incidents are violent, targeted at people (in
order to get their property), carry the threat of rape
and murder, and are often committed by large
groups of offenders. Compared to equally serious
violent crimes like rape, domestic violence and
child abuse, most of these robberies are committed
in public, and at times when people are most likely
to be present. As such, their confrontational nature
symbolises the lack of fear and apparent disregard
that many criminals have of being apprehended. All
these factors are what make robbery the key crime
problem for South Africa.
It is also notable that several types of aggravated
robbery, such as car hijacking and cash-in-transit
heists, have increased in the last reporting period,
contrary to the trend for robbery overall. These are
discussed in more detail below.
Murder
Between 1994/95 and 2005/06, the murder rate
decreased by 41% from 67 murders per 100,000
people to 39.5 (Figure 6). This is a particularly
positive trend for the country because murder
statistics are the most accurate on record, and also
provide the only reliable quantitative indicator of
the extent of violence in the society.
The murder rate has decreased steadily since
1995/96 by an average of 5% per year. The 2%
drop in the most recent financial year might suggest
a slow-down in the rate of decrease, although a
Figure 5: Aggravated robbery rate
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
100
150
200
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
250
300
350
220
196
164
177
220
230
260
262
279 288 272
255
N
o
p
er
1
0
0
,0
0
0
p
eo
p
le
SA CRIME QUARTERLY No 18 DECEMBER 2006 5LOUW
similar percentage decrease was recorded once
before (in 2002/03). It would be of concern if the
latest trend signifies the start of a levelling off of the
murder rate at the comparatively high rate of nearly
40 per 100,000 people.5
Key crimes that increased in the last financial year
Increases in three types of serious crime – car
hijacking, car theft and cash-in-transit robbery –
between 2004/05 and 2005/06 are noteworthy.
Car theft and car hijacking
Both categories of ‘car crime’ had, like the crimes
discussed above, been decreasing for some time up
to 2005/06 (Figure 7). After peaking in 1998/99 at
107,448 recorded incidents, car theft dropped
steadily until 2004/05 when 83,857 thefts were
recorded. Between 2004/05 and 2005/06, the
number of car thefts increased by 2.5%, to 85,964.
Car hijacking peaked a little later than car theft – in
2001/02 when 15,846 robberies were recorded by
police (Figure 8). Since then the number of
hijackings had declined steadily, reaching 12,434 in
2004/05. The last financial year however saw a
reversal of this trend, with a 3% increase noted
between 2004/05 and 2005/06.
One explanation for the recent increases may be
that more vehicles are being registered in the
country. It is also likely, given that it is largely
crimes that are typically organised that are rising,
that car theft and hijacking have increased together
with syndicated crime in general.
Cash-in-transit robbery
The trend for cash-in-transit (CIT) robberies drew
much attention when the crime statistics were
released a few months ago. Police reported that
these violent and highly organised attacks had
increased by 74% between 2004/05 and 2005/06 –
from 220 to 383 incidents (Figure 9). The increase is
certainly cause for concern especially considering
that only two years before, in 2003/04, the lowest
number of heists was recorded since 1996/97.
However, the CIT robbery trend has been erratic
since police started recording these crimes
separately in 1996/97, and it’s not the first time that
such high levels have been recorded: more than 300
incidents were recorded in 1996/97 (359) and again
in 2002/03 (374).
The substantial increase in CIT heists in the last
financial year may be partially attributable to a
change in the way the police define these robberies.
The South African Banking Risk Information Centre’s
(SABRIC) figures for CIT heists for the same period
show an increase of 28% (admittedly still a
significant percentage for one year).6 The
discrepancy could be the result of the police now
including what SABRIC call ‘cross-pavement
incidents’,7 which police had previously categorised
as general aggravated robbery rather than specific
CIT robbery. Given that SABRIC estimate that cross-
Figure 6: Murder rate
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
30
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
20
50
70
40
60
80
67 68
63
59
60
53
50
48 47
43
40
39.5N
o
p
er
1
0
0
,0
0
0
p
eo
p
le
1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
10,000
SA CRIME QUARTERLY No 18 DECEMBER 20066 LOUW
pavement robberies made up about 60% of CIT
heists in the last financial year, these incidents would
have a significant influence on the trends.
The increase in CIT robberies has been attributed to
various factors among which the following are
notable:
• the increase in cash in circulation;
• inadequate cash management processes;
• ineffective information sharing between affected
parties;
• the absence of minimum standards for vehicles,
training, vetting, etc. among the companies that
move cash;
• the fact that CIT guards are generally outnumbered
both in terms of personnel and firepower during
attacks;
• the limited use or availability of SAPS and backup
vehicles; and
• inadequate environmental design which results in,
for example, areas for cash pick-ups and drops at
shopping malls being the same as the areas
commonly used by the public.8
It is also interesting to note that the most dramatic
increases in CIT robberies occurred at the time
when the number of bank robberies started falling in
2001 (Figure 9). It is possible that the two trends are
related, with syndicates turning their attention to the
more vulnerable cash movement industry as banks
became higher risk targets. Even if the trends are not
related, the fact that the banking industry, together
with government, managed to secure an 89%
decrease in the number of bank robberies between
Figure 7: Vehicle theft
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
80,000
1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
100,000
120,000
105,867
98,669
97,332
102,571
107,448
103,041
100,030
96,859
93,133
88,144
83,857
85,964
N
u
m
b
er
Figure 8: Carhijacking
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
12,000
16,000
14,000
18,000
12,912 13,052
15,773
15,172
14,930
15,846
14,691
13,793
12,434
12,825
N
u
m
b
er
60,000
SA CRIME QUARTERLY No 18 DECEMBER 2006 7LOUW
experiences of crime during these six months differ
markedly from what the trends for the preceding
financial year show, the police run into a public
relations nightmare. To make matters worse, by the
time the figures are released, the public would have
been waiting a year and a half for the ‘latest’ statistics.
Responding to complaints about the release policy,
the Minister of Safety Security has said that next year
“the statistics will be published soon after the end of
the financial year and will be separated from the
SAPS annual report.”9 This is good news, as long as
the figures are made available no later than early May
of each year.
Police leadership’s response to the ‘crime wave’
The response of police leadership has left the public
with the sense that government – and the police in
particular – don’t care enough about the problem or
its consequences. The Minister of Safety and
Security’s remarks in parliament in June that those
who just “whinge” about crime (as opposed to doing
something about it) should leave the country, were
met with outrage from a cross-section of South
Africans – wealthy and poor, black and white.
In the wake of the Jeppestown incident and the other
violent robberies that followed, the police leadership’s
communication with the public was sporadic at best.
Occasional statements and media appearances were
1996/97 and 2005/06 is highly significant, and
illustrates that solutions can be found.
Official crime trends vs. public perceptions
Considering that the good news outweighed the bad
when the figures were released, why the heightened
interest in the 2005/06 crime statistics? The surge in
especially violent robberies over the past six months
noted above (which are not included in the latest
statistics) provide one answer. But current negative
sentiments about crime and safety are about more
than the media coverage of horrible incidents and
admissions from reliable sources about what the
statistics show.
SAPS policy on release of crime statistics
The SAPS release policy is partly to blame. Crime
statistics are released annually by the police, usually
at the end of September, to coincide with the release
of the SAPS annual report. The figures cover the
financial year, i.e. April – March. While an annual
release of crime figures is not theoretically a problem
(although more frequent releases are both possible
and desirable), the SAPS brings unnecessary attention
and pressure to bear on itself by releasing the crime
data six months after the end of the financial year.
This means that no figures are available for the six
month period between March and September, and as
events this year demonstrated, when public
Figure 9: Bank robbery and cash-in-transit heists
Source: Annual Report of the SAPS for 2005/2006, Pretoria, 27 Sept 2006.
0
100
200
300
400
1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06
500
600
561
463
493
450
469
356
127
54 58 59
359
236
223 226
196
238
374
192
220
383
N
u
m
b
er
Bank robbery Cash-in-transit heists
SA CRIME QUARTERLY No 18 DECEMBER 20068 LOUW
made by the Minister, but given the ‘whinging saga’,
he was probably not best placed to calm fears and
boost public confidence.
Throughout the 2006 ‘crime wave’, (with the
exception of media briefings to defend himself against
allegations of his links to murder-accused Glen
Agliotti) little has been heard from the national
commissioner of police. As leader of the SAPS – and
with the ability to control strategy, operations and
resources – Jackie Selebi holds the position which,
rightly or wrongly, symbolises government’s response
to crime.
What would help is a sincere and informed
acknowledgement of the current problem, followed
by a clear outline of specific responses to specific
crime problems and how these will be dealt with in
various parts of the country. Instead we have (from
the Minister) a brief two-page statement about a
‘strategy’ that covers generalised responses which
sound more like the day-to-day activities of the police
(and others) than a considered initiative to deal with a
crime wave.10 Silence from the national commissioner
on these issues is bound to raise doubts about the
police’s intentions and abilities, if for no other reason
than the absence of clear information to the contrary.
Of the information that has been provided to the
public, it is encouraging that organised crime has
been identified as a priority, and that, as a result,
intelligence capacity will be improved along with
border control. And although we still know little
about the precise nature of the SAPS restructuring
aimed at dismantling the Area level offices in order to
boost station level capacity, it is a relief that the SAPS’
specialised organised crime units won’t be affected. It
is also positive that the Minister has identified the
need to better understand violence in the country,
and has appointed the Centre for the Study of
Violence and Reconciliation (CSVR) to assist in this
regard.
Against this backdrop, it is not surprising that when
the official statistics were released by the Minister and
the commissioner, they were received with a mixture
of scepticism and alarm. The scepticism can be
attributed to the fact that the SAPS policy on the
release of data meant that the figures didn’t cover
events over the preceding six months that South
Africans were most concerned about. Even though
the release policy has been in place for some years
now, the timing of events encouraged general public
doubts about the accuracy of police crime figures.
The sense of alarm came in response to the police
leadership’s insistence that the news was
overwhelmingly good.
While there may well be merit in both the SAPS
release policy and the claim that the trends were
good news, the handling of public communications
by the police about crime information and policing
strategy is clearly out of sync with public sentiment
on the issues.
This is the context within which the latest crime
trends must be considered. And this is why – when
the overall analysis is that, up to March 2006, much
of the news is indeed good – anyone who has been
in the country over the past six months will remain
doubtful.
Endnotes
1 Sunday Times 2/7/06.
2 See A Altbeker, Positive trends: The 2004/05 crime
statistics, SA Crime Quarterly No 14, Institute for
Security Studies, 2005.
3 P Burton, A du Plessis, T Leggett, A Louw, D Mistry
and H van Vuuren, National Victims of Crime Survey:
South Africa 2003, ISS Monograph No 101, Institute for
Security Studies, Pretoria, July 2004, p 107.
4 See A du PLessis and A Louw, The tide is turning: The
2003/04 SAPS crime statistics, SA Crime Quarterly No
12, Institute for Security Studies, 2005.
5 For more on international comparisons of crime rates,
see A Altbeker, Puzzling statistics: Is South Africa really
the world’s crime capital?, in SA Crime Quarterly No 11,
Institute for Security Studies, 2005.
6 CIT Robberies, SABRIC media release, undated, 2006.
7 The cross pavement definition refers to all incidents
that take place when guards are accosted outside their
vehicles – typically whilst replenishing ATMs, or
picking up money from a merchant, cited in IJ van
Vuuren, Cash-in-transit robberies: A sudden escalation?,
paper presented at Institute for Strategic Studies
conference, May 2006.
8 Ibid.
9 Sapa and Carol Hills, Crime: ‘Major cause for
concern’, Mail & Guardian Online, 27 September
2006,
10 For details on the ‘strategy’, see