Conviction rates an unreliable benchmark of NPA success

Comments 7

South Africa's National Prosecuting Authority routinely boasts remarkably high conviction rates. It uses them to reject criticism of its performance. But as it only prosecutes cases it is likely to win, they are unreliable measures of success in tackling crime.

The conviction rate figures routinely cited by South Africa’s National Prosecuting Authority (NPA) and Department of Justice and Constitutional Development are impressive: 88.8% overall, 89.1% in organised crime cases, 65.1% in sexual offences prosecutions, 91.6% in complex commercial crimes and 96.1% in asset forfeiture cases.

Addressing a cyberlaw conference last week, Deputy Justice Minister Andries Nel boasted a 97.6% success rate in cybercrime prosecutions.

Last month, while speaking to secondary school pupils as part of a national Stop Rape Campaign, Nel made a similar claim of success in prosecuting sex crimes. During “the last financial year, the sexual offences units secured over 363 life sentences, with a conviction rate of 73 per cent for crimes against women above 18 years of age and 70 per cent for crimes against children under 18 years of age,” he said.

According to the NPA, conviction rates for so-called “trio crimes” – car hijackings, business robberies and house robberies – stand at 84.8%.

In cases of rhino poaching, trafficking of rhino horn and illegal hunting, the NPA has laid claim to an 83% conviction rate.

Are high conviction rates a sign of success?

At first glance, the numbers look like a remarkable achievement, but how accurate are they as a benchmark of performance?

Late last year, Nel used the figures to reject accusations that the NPA is “rudderless and in crisis”.

“[T]he NPA has achieved conviction rates of 84.6% in high courts, 73.3% in regional courts and 90.8% in district courts. This is not the performance of an organisation in crisis,” he said.

It depends on how you define the rate

The NPA defines a “conviction rate” as the “percentage of cases finalised with a guilty verdict divided by the number of cases finalised with a verdict.”

Convictions are counted at the date sentencing or “not guilty” verdict, irrespective of when a plea was first entered. Therefore a case instituted in 2011, but concluded in 2013 would form part of the 2013 conviction rate.

More to the point, conviction rates do not reflect the number of successful prosecutions in relation to the number of crimes reported to police each year, let alone the large number of crimes that go unreported.

Comparing prosecution and crime reveals more

Between January 2010 and July 2012, for instance, 573 people were arrested for rhino related crimes. Yet, in 2012, only 28 accused were convicted in 20 cases.

Figures released in February this year by the Gauteng  province’s Member of the Executive Council for Community Safety, Faith Mazibuko,  showed that 23,086 rapes were reported to police in the province between 2010/2011 and the third quarter of 2012/2013. Of that number, only 55.59% were referred to the NPA for prosecution.  The NPA, in turn, sent 8220 cases back to the police for further investigation. At least 8870 cases were thrown out of court due to incomplete investigations and only 1910 cases (about 8.2% of the total number reported in that period) resulted in successful convictions.

In the case of cybercrime, Nel’s claim of a 97.6% conviction rate was based on successful prosecutions in 123 cases.

The NPA last year argued that accusations that conviction rates are distorted are “simply unfair and disingenuous” and that conviction rates are “internationally viewed as an indicator of success in prosecution”.

Conviction rates ‘liable to abuse’: US study

But in a monograph published last year by the Institute for Security Studies, crime and justice researcher Jean Redpath noted that conviction rates as a measure of performance are “almost meaningless”.

She argued that the NPA’s focus on high conviction rates as a barometer of success has encouraged a “tendency to decline to prosecute”.

Prosecutors have a wide discretion to decide which cases have “reasonable prospects of success”.

Redpath used the example of a year in which the NPA received 517,000 case dockets from police. Just over 74,000, or about 14% of those cases, were ultimately prosecuted. The NPA declined to prosecute in more than 300,000 cases (60%) and referred a further 130,000 to police for further investigation.

Similarly, a US study on convictions versus conviction rates has found that “a prosecutor’s high conviction rate may not be a sign that he is tough on crime and doing a good job”.

“Instead, he might just be taking easy cases and letting too many criminals go without prosecuting them…”

“Any system which pays attention to conviction rates, as opposed to the number of convictions, is liable to abuse.”

Conclusion – without comparison to crime conviction rates meaningless

Understandably, the NPA only prosecutes cases it thinks it has a reasonable chance of winning. But, for this reason, the version of a conviction rate that it uses is, clearly, an unreliable test of both its success or lack of it, and that of the wider criminal justice system in South Africa.

Instead, the only accurate and sensible measure of performance of the criminal justice system would be a comparison of the number of successful convictions and the number of crimes reported to police on an annual basis. And there, the numbers are not so good.

Edited by Peter Cunliffe-Jones

© Copyright Africa Check 2020. Read our republishing guidelines. You may reproduce this piece or content from it for the purpose of reporting and/or discussing news and current events. This is subject to: Crediting Africa Check in the byline, keeping all hyperlinks to the sources used and adding this sentence at the end of your publication: “This report was written by Africa Check, a non-partisan fact-checking organisation. View the original piece on their website", with a link back to this page.

Comment on this report

Comments 7
  1. By Africa Check

    Our point is that conviction rate figures are not a reliable means of measuring the success of South Africa’s criminal justice system. They do not reflect the numbers of successful prosecutions in relation to the large numbers of crimes and arrests each year. The emphasis on high conviction rate stats also means that prosecutors are likely to avoid difficult cases and pursue only those where there is an extremely high probability of success. As the US study noted: “[A] prosecutor’s high conviction rate may not be a sign that he is tough on crime and doing a good job…Instead, he might just be taking easy cases and letting too many criminals go without prosecuting them…”

    Reply Report comment
  2. By Barry Sergeant

    The NPA’s rates are bogus. The Kebble frauds, now worth R30 billion, have, for instance, never been prosecuted; never mind any convictions. And so on, ad nauseum.

    Reply Report comment
  3. By Amanda Watson


    ‘According to a report in The Star newspaper, 23,086 rapes were reported in Gauteng in 2012. Of that number, 55.6% were referred to the National Prosecuting Authority to be prosecuted. But 35.6% of those were referred back to police for further investigation and 38.4% of the cases that were prosecuted were thrown out of court due to incomplete investigations.’

    This would be the 2011/2012 crime stats? Was this cross referenced with the South African Police crime stats?

    Also, do the numbers reflected in the statement refer only to cases reported in the 2011/12 timeframe, or all outstanding cases including those prior to 2011/12?

    Reply Report comment
  4. By Peter Lawton

    The tendency to use ‘percentage rate’ of claim something is indicative of nervousness or deviousness concerning absolute numbers. The NPA spokesperson is using a logical cascade of numbers eg ‘took 10 cases to completion, got 8 convictions, therefore 80% conviction rate achieved’; the logical cascade is fine. But is this a valid measure of their work? It can be, if both the percentage and the absolute numbers are given. A more honest and illuminating assessment would comprise regular and frequent reports containing sets of data that divulge the numbers of case dockets received, the number taken forward to trial, the number returned for re-investigation, the number not taken forward because of doubt about winning, the number of requests for deferred trial dates, the number of complaints to relevant authority about poorly assembled dockets/case-work received from investigators etc. This sort of transparency would earn respect. The ‘conviction rate percentage’ earns scorn from those who think. It’s use suggests disrespect for the intelligence of those who are supposed to be impressed by the ‘percentage trick’

    Reply Report comment
    • By Kathy Wiles

      Exactly Peter and the declining to prosecute is the very reason why crime grows, way too many criminals being let off the hook. I’ve had a personal and very bad experience with this myself where the Senior Prosecutor abused his power and treated me like dirt

      Reply Report comment
  5. By Africa Check

    Hi Amanda. Thanks for pointing that out. The newspaper report we referred to incorrectly conflated the figures to suggest that there were 23,086 rapes reported in Gauteng in 2012. We should have checked the numbers more closely.

    In fact, according to the response by the MEC, the 23,086 cases referred to were reported from the beginning of the 2010/11 financial year up to the third quarter of 2012/2013. We have updated the report accordingly and included a link to the MEC’s reply which provides a complete breakdown of the figures provided to her by police.

    Reply Report comment

Leave a Reply

Your email address will not be published. Required fields are marked *


Africa Check encourages frank, open, inclusive discussion of the topics raised on the website. To ensure the discussion meets these aims we have established some simple House Rules for contributions. Any contributions that violate the rules may be removed by the moderator.

Contributions must:

  • Relate to the topic of the report or post
  • Be written mainly in English

Contributions may not:

  • Contain defamatory, obscene, abusive, threatening or harassing language or material;
  • Encourage or constitute conduct which is unlawful;
  • Contain material in respect of which another party holds the rights, where such rights have not be cleared by you;
  • Contain personal information about you or others that might put anyone at risk;
  • Contain unsuitable URLs;
  • Constitute junk mail or unauthorised advertising;
  • Be submitted repeatedly as comments on the same report or post;

By making any contribution you agree that, in addition to these House Rules, you shall be bound by Africa Check's Terms and Conditions of use which can be accessed on the website.


This site uses Akismet to reduce spam. Learn how your comment data is processed.