Why the NPA is misfiring on high-profile corruption cases

Research released this week by the Africa Criminal Justice Reform (ACJR) unit at the University of the Western Cape shows that the number of corruption convictions against government officials has significantly declined in recent years.

For the 2020/21 financial year, the relevant conviction figures more than halved from where they were in 2016/17.

This slump seems to fly in the face of assurances from government leaders and law enforcement agencies that corruption is being taken more seriously than ever before. As ACJR researcher Professor Lukas Muntingh says: “The evidence simply does not support claims of a particular focus on corruption.”

The National Prosecuting Authority (NPA) says it is misleading to make this conclusion only on the basis of conviction numbers, and points to the impact of Covid-19, which has resulted in fewer cases being finalised.

“This is not an indication that the NPA has not prioritised these matters. A lot of work has taken place over the previous three years to enhance capacity and skills, and the fight against corruption has become even more of a priority for the NPA,” spokesperson Bulelwa Makeke told DM168.

The NPA was one of the institutions hardest hit by the Zuma years. As the research shows, those effects are still being felt in the large number of vacancies at the NPA, the dearth of highly skilled and experienced prosecutors as well as staff morale issues.

The NPA has also been accused of pursuing easy-win cases across all forms of crime at the expense of more complex prosecutions that could bring down its misleadingly high conviction rates. The researchers point to “a decline, if not near absence – bar a few exceptions, such as some commercial crimes courts – in complex but potentially high-impact cases being prosecuted”.

Worrying evidence

The research presented this week by the ACJR’s Muntingh and Dr Jean Redpath suggests some concerning trends in the NPA’s current performance, including:

• A decline in “overall throughput” (the number of cases in which a suspect is arrested, brought to trial, prosecuted and successfully convicted);


• A reliance on guilty pleas;


• A reliance on minor offences (particularly drug related) to beef up the numbers; and


• Large numbers of cases processed through the unregulated alternative dispute resolution mechanisms (for instance, when someone accused of assault agrees to pay the victim compensation to avoid a trial).

The NPA has been lauded for its high conviction rate – over 90% in recent years – but the researchers say this is a misleading metric by which to measure its actual performance.

It is problematic, Muntingh writes, as it provides an incentive for prosecutors to select cases “where there is a high likelihood of conviction and not pursue those where the chances are less favourable”. This approach is, arguably, particularly concerning for crimes such as sexual violence.

Sheena Swemmer, the head of gender justice at the Centre for Applied Legal Studies, told DM168 that the unit is concerned about the “really huge number of rape cases” where the NPA declines to prosecute.

“The ‘perfect rape case’, where there is a winnable case, is usually what they are looking for,” she said. “They will shy away from cases where women were drunk or under the influence of drugs.”

The decision to prosecute “is really determined by how difficult it might be to disprove a rape myth, or perhaps influenced by the belief in rape myths”.

The NPA’s Makeke acknowledged that “the prosecutor must be satisfied that there is a reasonable prospect of success”, but says that this is guided by whether the available evidence meets the required standard. The use of conviction rates to measure NPA performance is also misleading, the ACJR researchers note, because it only reflects the cases the NPA decides to prosecute, and not those the NPA chooses not to proceed with. Redpath says that, in the early post-1994 years, the conviction rate was measured “as a proportion of cases opened by the SAPS [SA Police Service] and not, as is currently the case, as a proportion of cases prosecuted”.

The data analysed by the researchers show that, by 2020, only about half of the cases regarded by the police as amounting to sufficient grounds for arrest were ultimately viewed by the NPA as worthy of prosecuting.

Makeke said it was “not possible” for the NPA to determine the number of arrests that result in or do not result in a prosecution.

Until 2017, the SAPS counted the number of convictions stemming from charges, tracking the number of cases making it all the way through the criminal justice system. Between 2010 and 2017, writes Redpath, “there was a collapse in the conviction throughput rate” for all crime types except drug-related offences and aggravated robbery.

The SAPS no longer includes conviction numbers in its annual reports so it is now even more difficult to measure NPA performance beyond what Redpath terms the “totally inappropriate indicator” of conviction rates.

What the NPA could do

When the heads of three vital criminal justice bodies – the NPA, the Hawks and the Special Investigating Unit – appeared before Parliament this week, there appeared to be one major question on the minds of MPs: Why isn’t South Africa seeing more action on prosecuting high-profile cases?

DA MP Robert Lees summed up the concern, saying: “What South Africans are looking for is significant numbers of high-level successful prosecutions and convictions.”

The ACJR research makes it clear that the NPA is still beset by very real internal challenges, many relating to staffing. The data show that between 2014 and 2020, the number of prosecutors employed by the NPA declined by 22%. Staff surveys in 2019 and 2020 showed that the biggest sources of discontent for NPA employees were the lack of promotion opportunities, the high vacancies, budgetary constraints, a perceived lack of leadership support and low staff morale.

The NPA struggles to hang on to talented prosecutors in these circumstances. Without these skills and experience, it becomes very difficult for the NPA to build strong cases.

But it has also been proposed that, in the face of its obvious limitations, the NPA should refocus its efforts in simpler ways.

Redpath has argued that, in cases related to State Capture, for instance, the NPA should “not feel compelled to create complicated cases by bringing together all matters of which a heavily implicated person is likely guilty”. One example is that of former SAA chair Dudu Myeni, whom State Capture reports have recommended for potential prosecution for offences related to State Capture.

Start by prosecuting Myeni for contempt of court, Redpath and others have argued. Myeni named a protected witness during the State Capture inquiry against the explicit instructions of Judge Raymond Zondo.

“A single such prosecution may signal serious intention on the part of the NPA, and encourage other accused to agree to plea-and-sentence agreements,” Redpath contends. It is always possible to charge Myeni with further offences down the line.

The ACJR researchers also called for far greater transparency from the NPA.

The Centre for Applied Legal Studies' Swemmer said, for example, that complainants seeking to understand why their rape case would not be prosecuted often had to launch a Protection of Access to Information Act (Paia) request to the NPA to hear those reasons. The NPA denies this, with Makeke saying: “Where the NPA declines to prosecute, reasons for these decisions are provided to the affected parties.”

If the NPA made speedier decisions about which cases not to prosecute, and disclosed this publicly, it would also make it more feasible for private prosecutions to be brought in a timely manner.

NPA EXCUSES – AND RESEARCHERS’ REBUTTALS

Excuse: Despite its clear lack of capacity and technical skills in areas such as forensic accounting, the NPA has said it is reluctant to accept assistance from the private sector or civil society in prosecuting State Capture cases to avoid accusations of bias.

Rebuttal: Dr Jean Redpath pointed out in an article in February that it is the decision to prosecute – and not the building of the case – that is open to bias: “It is difficult to see how independence is compromised in Zondo-identified cases, where the most senior judge in the land has already found that there is, in NPA parlance, a prima facie case and clear public interest in prosecuting.”

Excuse: NPA head Shamila Batohi told Parliament this week that one reason for the agency’s slow progress was budgetary challenges.

Rebuttal: Redpath writes in a new paper that the NPA’s funding fluctuations since democracy cannot easily be correlated with its success or failure. She states that, even allowing for inflation, the NPA obtained “a far higher number of serious crime convictions” with lower budgets than currently.

Excuse: The NPA has said that, even when a seemingly strong case is referred from the Special Investigating Unit (SIU), it could require much more work before it is ready for court.

Rebuttal: Redpath suggested in her February article that the NPA should take advantage of the provision in law allowing for the SIU to carry out its own prosecutions. “The SIU has lawyers familiar with most of these matters,” wrote Redpath; “the SIU can hardly do any worse than [the NPA has] done.”

SOME OF THE (MANY) CASES FOR WHICH SA HAS WAITED TOO LONG

• The murder of the Cradock Four: The NPA was handed evidence relating to the 1984 murder of the anti-apartheid activists known as the Cradock Four by the TRC in 2003. No prosecution on this or many other TRC matters has happened, though the NPA has promised a renewed focus in 2022.


• Steinhoff: The company’s fraudulent accounting practices were made public in December 2017. Alleged mastermind Marcus Jooste has yet to see the inside of a local court.


• Mosebenzi Zwane and other Gupta enablers: There was already clear evidence from the 2017 GuptaLeaks stories, let alone the State Capture inquiry, that former Cabinet ministers such as Mosebenzi Zwane bent over backwards to facilitate Gupta corruption. Where are those prosecutions?


• On-Point Engineering: EFF leader Julius Malema and others were charged with more than 50 counts of fraud and corruption in 2012 over a construction firm part-owned by Malema’s family trust, which is accused of winning kickbacks on contracts to fix Limpopo’s failing infrastructure. The case was struck from the roll in 2015 after too many state delays. Some charges were reportedly reinstated against Malema’s co-accused in 2019, but inexplicably not Malema.


• Pandemic procurement looters: The SIU has referred 386 cases relating to Covid-19 procurement corruption to the NPA. Some of these – such as the involvement of former Health Minister Zweli Mkhize – were exposed by the media months before.


• The instigators of the July 2021 unrest: Some arrests have been made for incitement of the July 2021 violence on social media, but South Africa still awaits meaningful prosecutions of those suspected of being behind the unrest.


• The Malema firearm case: In July 2018, Malema was filmed firing an assault rifle during an EFF rally. Surely an easy-peasy case, with video evidence to hand: the law is clear that it is illegal to discharge a firearm in public. After more than three and a half years, Malema was only just pleading in court by March 2022. We could fill this list with further, far more serious, unresolved Malema matters – not least the allegations relating to the collapse of VBS Mutual Bank. DM168

This story first appeared in our weekly Daily Maverick 168 newspaper which is available for R25 at Pick n Pay, Woolworths, Spar, Exclusive Books and airport bookstores. For your nearest stockist, please click here.