|
|
South Africa | Change |
|
|||||||
|
|||||||||
| Contact us |
|
|
Special Report: Crime data |
| 8 Oct 2009 | |
Crime data in South Africa has, in the past, been an emotional topic, with analysts and reporters finding it difficult to qualify the actual numbers released. This problem is partially due to the emotional impact of such data on South African society. It is therefore desirable to compile an accurate, objective measure of crime trends and levels, but two other obstacles are clear in the data itself.
Common problems in interpreting crime data
The first obstacle is related to measuring crime trends. Although it is desirable to know whether, on average, crime is increasing or decreasing within the country, it is almost impossible to objectively make such conclusions based on the released raw data. This is because, although one category of crimes may show an increase, another may in fact show a decrease. Understanding the overall trend is further complicated by the fact that not all crimes are equal.
If, for example; shoplifting were to decrease by a given percentage and murders increase by the same percentage - we cannot say that society is in an equal position. Clearly, the decrease in crime due to less shoplifting does not offset the increase in crime caused by more murders. In fact, even if shoplifting were to drop by 300%, and murders increase by only 10%, one could easily argue that crime has in fact increased due to the more destructive nature of murder over shoplifting.
The second problem is a much more common, and more easily solved one; that of population growth. If, for example; the incidences of burglaries remained the same from one year to the next, but the population increased over the same period, then clearly the rate of burglaries has decreased. Even though there were more people to burglarise, and more potential burglars, the number of burglaries remained constant. We can therefore say that, per person, there were less burglaries, and therefore less chance that a randomly selected person would have been burglarised during that year. In other words, people were, on average, less likely to be burglarised from the one year to the next and that society is better off. Crime rates have decreased.
The IHS Global Insight Crime Index
The above problems complicate the comparison of one region to another, particularly as different trends and differently sized populations are involved. Therefore, in order to better understand the crime statistics on a regional level, we have developed the IHS Global Insight Crime Index. This is useful not only for meaningful regional comparisons, but also for comparisons over time and for spotting trends in overall crime.
Methodologically, all crimes are adjusted for population, and then weighted according to the seriousness of each offence. They are finally combined into a single index figure which is useful for comparisons over time and area. The completed crime index makes useful analysis of the crime data possible.
National crime levels
Of immediate interest to most observers is the level and trend of crime for the country as a whole. The common perception on South African crime is one of an increasing crime environment. This perception is not borne out in the actual data. Such perceptions may be due to media reports that focus on one type of crime, which may be increasing, and neglect another type of crime, which may in fact be decreasing. Furthermore, individual perceptions are subjective and easily biased by personal experiences of crime.
However, despite the common perception, South Africa has actually been in a period of decreasing overall crime since 2003, when the overall crime index peaked at 171 (right axis on following chart.) In fact, the overall level of crime is currently 13.6% lower than it was at the start of 1997, and 24.7% lower than its highest peak since then.
Certainly, the absolute number of murders (the most serious crime category and therefore the one with the highest weighting in the index) has dropped from 25,470 in 1996/97 to the most recent 18,148.
Source: IHS Global Insight Regional eXplorer (ReX)
Looking at the latest crime data release, the overall level for 2008/09 is almost the same as it was for the previous year. In other words, the average citizen was just as likely to be a victim of crime during 2008/09 as they were during 2007/08. In reality, the overall crime index is ever so slightly higher, up 0.23% on the previous year. It may be useful to look at this rate of change variable more closely.
Source: IHS Global Insight Regional eXplorer (ReX)
From the above chart, we can see that crime has been in a state of increasing decline since around 2003. However, since the 2006 data release, crime has been decreasing at a slower and slower rate, and with the release of the latest crime statistics, property crime has in fact started to rise again.
There is no doubting the serious nature of crime in South Africa. With the latest statistics declaring a murder rate of 37.58 (per 100,000) and a total sexual crimes rate (which includes rape, indecent assault and abduction) of 146.02. Nonetheless, the declining rate of crime is encouraging. However, at the current pace of crime reduction, it would take between 6 and 18 years to bring the South African murder rate down to that of comparable countries with ‘low' murder rates.
Crime and World Cup 2010
Although the majority of violent crime perpetrators are known to their victims, the negative international perception that South Africa has regarding its high crime levels could potentially hurt tourism. It may be interesting, therefore, to compare crime levels across the FIFA World Cup 2010 host cities and the national average.
Source: IHS Global Insight Regional eXplorer (ReX)
The above chart shows the level of crime in the 2010 host cities as compared to each other, and to the national average. We can immediately see that most urban areas have a higher crime rate than the national average. It is very common for crime rates to be higher in urban areas than in non-urban areas, and South Africa is clearly no exception. Of interest however, is that overall crime has had an ambiguous trend across the host cities since 1996, with some cities experiencing increased crime and others decreased crime. However, the host cities of Port Elizabeth and Polokwane have experienced crime decreases larger than the national average of 13.65%.
Polokwane has actually experienced an over 50% drop in overall crime levels in the last 12 years. In absolute terms, the murder rate in Polokwane has dropped from 34.52 (per 100,000) in 1996/7 to 16.90 in 2008/09. Johannesburg and Bloemfontein both share the highest crime rate of all host cities.
It may be useful to extract only violent crime from the above. Violent crime is certainly a more powerful deterrent to tourists than property crime. Violent crime is often called ‘contact crime' or ‘body' crime as it includes all crimes that bring physical harm to the victim. With regards the IHS Global Insight crime index, violent crimes typically have a higher weighting than do property crimes. This is due to the added cost on society of emotional trauma, the lost productivity caused by such contact crimes and the overall destructive nature of the crimes themselves.
Source: IHS Global Insight Regional eXplorer (ReX)
Evidently, a similar picture emerges as that of the one above, with Polokwane emerging as the least violent host city, and Johannesburg and Bloemfontein as the most violent, in terms of contact crimes.
In short, although crime is certainly decreasing in the 2010 World Cup host cities, it is doing so at a marginally slower rate than the national average. Absolute levels of crime are also higher in the host cities than the national average. The fact that crime rates internationally are typically higher in urban areas than in rural ones is borne out in the South African data as well.
Provincial Crime
It may also be useful to have a quick look into crime on a provincial level. There is a large variance between crime rates in the different provinces. As expected, Gauteng and the Western Cape experienced the highest rates of overall crime in the 2008/09 financial year. This is likely due to their larger populations, their higher levels of relative poverty, a larger gap between the rich and the poor and their higher urban populations. Limpopo, once again, experienced the lowest levels of overall crime, fairly below the Eastern Cape (3rd lowest) and Kwazulu-Natal (2nd lowest.)
Source: IHS Global Insight Regional eXplorer (ReX)
Accuracy of crime statistics
There has been some debate over the accuracy of crime statistics in South Africa. In particular, there have been accusations of crime being underreported by police stations. The immediate question is; how much of an impact would this have on the IHS Global Insight Crime Index? We believe that impact to be very small.
If a station was, in fact, underreporting crime levels, it would make a lot of sense to focus on high volume crimes that go easily unnoticed. For instance, murder would not be such a type of crime, with an average of only 16 incidents reported per police station per year. Crimes like theft out of motor vehicles or all crime not mentioned elsewhere probably would. In general, the more serious a particular crime, the more difficult it is to underreport or ignore.
The IHS Global Insight Crime Index places a higher weighting on more serious crimes, and a lower weighting on less serious crimes. Underreporting of the latter will have an exponentially lower impact on the overall index. In essence, the easier it is to underreport a crime, the lower the impact that crime has on the index. Crimes that are the most difficult to underreport are the most important to the overall index level.
To illustrate this point. Even if we were to assume that incidences of theft out of motor vehicles were severely underreported, by 50%. (A figure of 50% would be unrealistically high but serves as a good demonstration.) The total effect on the overall crime index would be smaller than 0.8%.
Feedback
Please send any suggestions or feedback to info@ihsglobalinsight.co.za or contact IHS Global Insight on +27 (0)12-665-5420 if you have any queries.
The IHS Global Insight Crime Index, the absolute crime figures and the crime rates are available for all local municipalities and magisterial districts in the IHS Global Insight Regional eXplorer.
| For accurate and up-to-date economic, socioeconomic, demographic, and development information on a spatial level for South Africa, click here |
| Gerhard Bijker Product Manager - Regional eXplorer | |
| Phone: | +27 12 665 5420 |
| Email: | gerhard.bijker@ihsglobalinsight.co.za |