Back to Africa Check

How to read South Africa’s quarterly crime statistics after the Covid pandemic

South Africa’s latest quarterly crime statistics show sustained increases in several violent crime categories. The numbers reveal where and how these crimes happen, but to assign the right resources to the right areas we must understand the data and trends in context.

Numbers and statistics are essential for understanding crime. They tell us how many crimes were recorded during a certain period, the types of crime, and where they took place (or, at least, the police station where they were reported). 

Using this data we can calculate crime “rates” – such as how many crimes take place in an average day, month or year, or how many crimes occur based on the population of the precinct, province or country. 

If we have the stats from previous months and years, we can also look at whether crime numbers or rates have changed over time. When we compare incidence of crime across longer periods of time, or between different areas, it’s important to use crime rates per person, as population figures change over time and vary between places. For example, Gauteng province has high numbers of reported crimes but also the largest population in the country – more than a quarter of all South Africans live in Gauteng.

When we want to compare changes in crime over a short period of time, or within the same area, we can compare both raw numbers and crime rates to understand short-term shifts and local trends. This is one reason why it’s so important to have regular, and regularly updated, crime statistics. Until fairly recently, the South African Police Service (SAPS) only provided new crime statistics once a year, and for the previous reporting year. 

This meant that by the time the information was released some of it was more than a year and a half old. This was okay for comparing long-term trends but less helpful for spotting and addressing crime issues as they were happening. 

In 2020, the SAPS started releasing quarterly crime statistics, for three-month periods. This allows us to get closer to real-time insight into crime in South Africa. But reading these numbers still requires accuracy and context. 

The problem of inaccurate and out-of-context crime stats 

News reporting on crime stats in the first years of the Covid pandemic highlighted why both accuracy and context are important. 

In April 2020 police minister Bheki Cele announced that 87,000 complaints about gender-based violence (GBV) had been received in a single week – a claim widely and prominently repeated by media. Except, as Africa Check revealed, the correct figure was 2,300 complaints during the first week of lockdown in March 2020. The 87,000 was the total number of GBV complaints for the whole of 2019.  

In August 2021, when SAPS released crime statistics for the first quarter (Q1) of that year, news reports headlined claims that violent crime had skyrocketed, with infographics showing, for example, that murder had increased by a staggering 66% over the same quarter of the previous year. 

This information was accurate, but it was misleading because it compared crime statistics from Q1 of 2021 with those of Q1 of 2020. In the latter period South Africa (like many other countries) was under strict lockdown conditions that severely restricted people’s movement – and which saw drops in all forms of reported crime, most particularly in so-called contact crime. News reports were comparing the same region over a similar time frame, but it was an unsuitable comparison because of the vastly different circumstances in each year. 

What’s also clear in these examples is the important role the media plays in reporting crime. News platforms have come a long way since I wrote Africa Check’s guide to crime statistics in South Africa nearly a decade ago. Today, newspapers and news sites often produce credible graphs, infographics and analyses of the latest crime data. Many of these are helpful in visualising and understanding crime incidents and trends at a glance. 

But South African media is still prone to emphasising sensationalist – not moderate – data. As my own work researching violence and media demonstrates, this contributes to skewed perceptions of crime and risk that don’t always match the reality of where and how crime happens, or who it affects. 

A worrying increase in violent crime 

The latest crime statistics, for the second quarter of 2022 (July to September), are an important benchmark for South Africa. This is the first time since March 2020 that we have successive crime statistics not influenced by lockdowns, curfews and alcohol bans. 

But comparing the latest statistics to the same period in 2021 may still be misleading, as in that year South Africa still had some restrictions to control the spread of Covid. So I prefer to look at reported figures for 2018 and 2019 when comparing crime numbers between years.   

For the last decade, violent crime in South Africa has been on the increase. The 2022 crime data shows that not only have many crimes recovered to pre-pandemic levels, but in a worrying number of categories they have started to exceed the numbers recorded in 2019, the last year before the pandemic. 

This is mainly notable in cases of murder, attempted murder and robbery with aggravating circumstances. Sexual offences have also shown increases, although these are less reliable given the vast under-reporting of sexual offences.

Crimes reported for July, August and September 

Crime

July to Sept 2018

July to Sept 2019

July to Sept 2020*

July to Sept 2021**

July to Sept 2022

Murder

5,554

5,446

5,107

6,163

7,004

Attempted murder

4,889

4,856

4,941

5,157

6,155

Robbery with aggravating circumstances

36,304

37,617

32,745

31,480

38,412

Sexual offences

12,745

13,730

11,423

11,964

13,283

Total number of reported crimes (South Africa, all provinces). Source: SAPS
* Includes periods of strict lockdown and total bans on the sale of alcohol.
** Includes some restrictions on movement and the sale of alcohol

Murder

South Africa is currently on track to post its highest murder numbers in 20 years. This requires more than just general alarm, as a closer look at the statistics highlights particular regions as being mainly responsible for the jump. 

Almost all provinces show increased numbers of murders per person, but the highest increases are in the Eastern Cape and KwaZulu-Natal. Both provinces had relatively high murder rates even before the pandemic, and the rise in murders in KwaZulu-Natal may be linked to the aftermath of July 2021 riots, and even the April 2022 floods. (Natural disasters can fracture unstable social structures, resulting in interpersonal violence.) 

Similarly, the rise in murders in the Eastern Cape might be linked to the province’s disastrous response to the Covid pandemic – it had, by far, the highest number of excess deaths per person in the country. This in turn speaks to poor political management, resulting in collapsing health systems and other essential infrastructure, problems that long predate Covid.  

Gauteng has also seen a notable rise in the number of murders, although the steepness of the growth is offset by the province’s large population. This, together with growth in other types of contact crime, suggests a broader issue with criminal violence. 

Issues of rapid urbanisation or in-migration, and unemployment as a result of the pandemic, may have played a role. Gauteng (and the Western Cape) experienced the highest levels of in-migration during Covid, as people moved to urban areas in search of work. This type of disruption can contribute to social instability and volatility, which would be compounded by the country’s extremely high unemployment rate. 

Other concerning trends revealed in the latest stats include continued growth in the number of mass killings and in the share of murders committed with firearms. Gun ownership is a hotly contested topic, but research has consistently shown that tighter gun controls reduce murders, and that even the presence of a firearm during a robbery increases the likelihood of someone being killed.

Although femicide is not a distinct legal or criminal category, police figures for the murders of women also indicate South Africa is on track to a possible record-breaking number of femicides – well above 3,000 female homicides, if current trends hold. If the femicide rate shows a higher rate of increase than the overall murder rate, this would be cause for specific concern. 

Other contact crimes of concern: hijacking and kidnapping

Gauteng typically records the highest number of contact crimes in the country. This is because it’s the most populous province – and a major economic hub.  In simple terms, there’s more to steal. Data from Gauteng provides an interesting barometer on crimes like robbery and car hijacking. 

Robbery requires contact between perpetrator and victim – if there’s no contact then it’s counted as theft – which creates a greater risk of violence. Recent trends in Gauteng’s data show that, while home and business robberies are close to or even slightly below pre-pandemic levels (in other words, high but not higher), carjackings and truck hijackings have jumped since 2018 and 2019.

Robbery and hijacking in Gauteng

Crime

July to Sept 2018

July to Sept 2019

July to Sept 2020

July to Sept 2021

July to Sept 2022

Carjacking

2,021

2,320

2,435

2,518

2,945

Robbery (residential)

2,166

1,903

1,942

1,829

1,906

Robbery (non-residential)

1,816

1,790

1,552

1,573

1,576

Truck hijacking

157

163

248

253

335

(Source: SAPS)

This has a direct connection to another crime category that’s shown steep increases, and often makes headlines: kidnapping. More than 40% of kidnappings across the country, and nearly 55% of all kidnappings in Gauteng, are linked to vehicle hijacking. (The second-largest number is linked to robbery.) 

This suggests a worrying change in the methods of carjackers and robbers. It requires an urgent response not only from the SAPS but also from private security companies. Commercial solutions like digital surveillance tools (cameras and so on) are not going to be much good in preventing kidnapping.  Millions have been spent installing privately owned CCTV cameras across wealthy suburbs, but these have likely done little except enrich suppliers. 

It should also be pointed out that cases linked to alleged human trafficking make up just over 0.5% of reported kidnappings. The media and NGO lobby groups emphasise trafficking as a present and rising danger, but there is no evidence that this is so in the vast majority of cases.  

Crimes detected as a result of police action 

“Crimes detected as a result of police action” are usually arrests resulting from things like roadblocks looking for drunk drivers, drugs and ammunition, or due to other coordinated police action. This category dropped by over 30% in the second quarter. 

We might expect arrests for, say, drug possession to decline (due to draft changes in the law against cannabis possession and use), the data suggests that the drop is largely because of reduced police operations and not due to an actual reduction in crime. So: fewer roadblocks (which we might all celebrate), but also fewer in-person, on-foot police operations. More questions about police deployment need to be asked for us to better understand these numbers. 

Dr Nechama Brodie is acting coordinator of the Wits Justice Project, and a lecturer at the Wits Centre for Journalism. Between 2013 and 2018 she worked with Africa Check, as a researcher and as the head of training and research unit TRI Facts. Her most recent books are Femicide in South Africa and Farm Killings in South Africa, both published by Kwela Books. 

This analysis is published as part of a collaboration between Africa Check and the Wits Justice Project that seeks to increase understanding around issues of crime and justice in South Africa.

Republish our content for free

Please complete this form to receive the HTML sharing code.

Further Reading

Add new comment

Restricted HTML

  • Allowed HTML tags: <a href hreflang> <em> <strong> <cite> <blockquote cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd> <h2 id> <h3 id> <h4 id> <h5 id> <h6 id>
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.
limit: 600 characters

Want to keep reading our fact-checks?

We will never charge you for verified, reliable information. Help us keep it that way by supporting our work.

Become a newsletter subscriber

Support independent fact-checking in Africa.