SASSA Grant Reviews: The Truth After 20 Years

SASSA grant reviews got two very different verdicts within days of each other. One came from SASSA itself. The other came from a pensioner queue.

Two opinion pieces, published four days apart, reaching opposite conclusions about the same programme.

SASSA grant reviews turned into a genuine public argument this week, playing out not in Parliament but across two opinion pages. SASSA marked its 20th year of operation with a piece on SABC News defending its review programme as essential, disciplined, and legally required. Four days earlier, IOL’s The Post had published a very different account, one built around a pensioner in a queue that started before sunrise, holding an SMS she didn’t fully understand.

Both pieces are, in their own way, right about something. SASSA pays approximately R20 billion a month to more than 19 million beneficiaries, a programme of a scale that genuinely does carry, in the agency’s own words, an enormous moral, legal and fiscal responsibility. The counter-argument is also right that the agency asking beneficiaries for financial discipline runs an uncomfortable history of its own leadership failing exactly that standard.

This article lays out both cases properly, adds the Treasury paper trail that explains why SASSA grant reviews exist in their current form, and separates the parts of this debate that are genuinely contested from the parts where the record already speaks for itself.

The Official Case: Why SASSA Says Reviews Matter

SASSA’s own framing, published to mark two decades of operation, is straightforward. Grant reviews exist because the Social Assistance Act of 2004 requires them, not because the agency has decided beneficiaries are suspects. People’s circumstances change: someone finds work, a household’s income shifts, a marital status changes, and the law requires SASSA to periodically confirm a beneficiary still meets the criteria for the grant they receive.

SASSA is also explicit that a review notice is not a punishment. Being flagged means additional verification is required before a final decision is made, not that a decision to remove someone has already been taken. This publication has covered that distinction in detail before, including the document checklist that determines whether a review appointment goes smoothly.

The technology behind this claim is real and expanding. Biometric verification now links to the Department of Home Affairs in real time, e-Life Certification lets beneficiaries confirm eligibility digitally, and SASSA argues this combination is what makes reviews less burdensome than they were even two years ago, particularly for older beneficiaries, people with disabilities, and those living far from an office.

SASSA’s anniversary framing also leans heavily on scale as justification. Twenty years ago, the agency did not exist in its current form; provinces ran their own, inconsistent grant systems, with wildly different administrative standards and no unified way to catch a person claiming the same grant twice across provincial lines. Centralising that system under one agency, SASSA argues, is precisely what makes a modern, biometric-linked review process possible in the first place, and precisely why abandoning verification now would undo two decades of consolidation.

Where the Pressure Actually Comes From

Neither opinion piece dwells on this part, but it explains a great deal about why SASSA grant reviews look the way they do in 2026. National Treasury attached specific, binding conditions to SASSA’s budget allocation letter: the agency must run bank income checks on multiple grant categories, conduct large-scale database checks at least twice a year, finalise data-sharing agreements with SARS and NSFAS, and report quarterly on how many grants were reviewed, suspended or cancelled.

This is not a policy SASSA dreamed up independently. It is a funding condition. Parliament’s own MPs have said as much, voicing concern about how the new verification requirements were landing on poor households even as they acknowledged SASSA had little choice but to comply. The review programme exists, in its current intensity, because Treasury made continued funding contingent on it.

How the Savings Claims Have Grown

Figures as reported to Parliament and in ministerial statements across the review programme’s rollout.

The trajectory matters. Each successive figure is larger than the last, not because fraud is accelerating, but because the verification net has widened: more bank accounts checked, more credit bureau records cross-matched, more grant categories included. This publication has previously reported the underlying scale, six million bank accounts and eight million credit records checked, 291,000 grants flagged, over 34,000 cancelled, and the trend in this table suggests that scale is still growing rather than levelling off.

The Counter-Case: A Pattern, Not Just a Policy

IOL columnist Sanjith Hannuman’s critique does not dispute that SASSA has a legal duty to run reviews. His objection is to how the process performs in practice, and to who is being asked to trust it.

He cites SASSA’s own mid-April figures: of roughly 15,500 people who used the online verification portal, 88% were verified successfully, but facial recognition alone generated close to 7,800 complaints against almost none for fingerprint checks. A beneficiary who fails the face-scan is sent back into the same physical queue the agency’s 1,000 new contract workers are meant to be relieving.

Black Sash, the advocacy group that has monitored SASSA for decades, has gone further, arguing the review process risks stripping more than 200,000 people of grants they still qualify for, through bulk SMS notifications that mean little to a beneficiary with no airtime or data. Their framing is blunt: a review without a fair opportunity to respond is “just a suspension with paperwork attached.”

Distance compounds the problem. This publication has separately documented a Northern Cape beneficiary living more than 20 kilometres from his nearest office. Hannuman’s column cites a comparable case, a pensioner 40 kilometres from her office with no way to save for a taxi fare on short notice, describing beneficiaries being pushed toward informal lenders simply to comply with a review they had no real way to prepare for.

The deeper structural criticism in Hannuman’s piece is about who carries the burden of proof. A working-age professional whose bank flags an unusual transaction gets a phone call and a chance to explain before anything happens to their account. A grant beneficiary flagged by the same kind of automated system gets a bulk SMS and a countdown, with the practical consequences of non-response landing well before any human being has reviewed the specific circumstances. The technology is broadly similar. The margin for error afforded to the person on the receiving end is not.

The R4.8 Billion Question, Again

This publication reported earlier this year that the Department of Social Development returned R4.8 billion in unspent grants funding to Treasury in a single financial year, R793 million of it sitting unused in the Old Age Grant alone, at the same time officials were tightening the very verification rules meant to protect that budget. Hannuman’s column raises the same figure, doing the same arithmetic this publication did months ago: split across beneficiaries, that underspend works out to roughly R1,200 extra per person for the year.

The pointed question both pieces circle, from different directions, is the same one: an agency that says money is too tight to avoid tightening verification on pensioners was, in the same period, sitting on billions it never paid out. SASSA’s official case does not address this directly. It is the strongest single argument in the counter-case.

Fraud Is Real Too: The Cases on Record

It would be a mistake to read the counter-case as a claim that fraud is invented. This publication has already reported one entire province’s worth of confirmed cases: 14 officials dismissed in the Eastern Cape for fraudulent grant approvals and unlawful bank detail changes, part of a national total of 43 dismissals in the past financial year.

Further cases have surfaced since. Four officials at SASSA’s Nebo office were dismissed over a R33 million scheme run with outside syndicates. A beneficiary in the Eastern Cape was sentenced to six months in prison for submitting a fraudulent medical assessment. A woman in the Northern Cape was arrested over an alleged loan racket run off other beneficiaries’ grant cards.

One widely feared claim has also been partly walked back. SASSA says most of the roughly 74,000 flagged “ghost beneficiary” accounts, deceased individuals apparently still receiving payments, turned out to be reporting-timing mismatches rather than actual theft, with no payments genuinely released after death in the majority of cases. That is a meaningful clarification, and it cuts against the more alarming version of the fraud narrative that circulated earlier in the year.

It is worth holding both facts at once, because the two opinion pieces each tend to foreground only the one that supports their argument. Fraud cases with real names, real dismissals and real prison sentences exist on the record this year. So does a large administrative flag that turned out to be mostly noise rather than theft. A programme this size will generate both outcomes simultaneously, and neither one, on its own, settles whether the overall balance of harm and benefit currently favours beneficiaries or works against them.

The History That Makes Beneficiaries Wary

Trust in this system does not start from zero, and it does not start clean. In 2018, the Constitutional Court found then-minister Bathabile Dlamini personally “reckless and grossly negligent” over the Cash Paymaster Services contract crisis that nearly cut off ten million beneficiaries, and ordered her to personally pay 20% of the legal costs, the first time a public official had been held personally liable in that way.

This publication has separately reported on Minister Pule’s own history in detail: her 2013 dismissal as Communications Minister after Parliament’s ethics committee and the Public Protector found she had concealed a relationship with a businessman whose companies benefited from a conference budget that grew substantially beyond its original allocation. She now leads the department running the exact kind of financial-disclosure crackdown that her own conduct once failed to meet.

None of this proves the current review programme is being run in bad faith. Pule was formally sanctioned, has publicly framed her return as an opportunity to rebuild trust, and the ANC’s own defence of her appointment rested on the argument that accountability for past conduct had already run its course. But it does explain, more plainly than either opinion piece states outright, why a beneficiary reading a review notice signed off under this ministry might reasonably want more than a press statement before extending the benefit of the doubt.

Where That Leaves You, Practically

Common Questions About the SASSA Grant Reviews Debate

Q1. Are SASSA grant reviews legally required, or a policy choice?

Both. The Social Assistance Act legally requires periodic eligibility checks, and National Treasury has additionally made continued SASSA funding conditional on specific, expanded verification measures.

Q2. Why does facial recognition fail so much more often than fingerprint checks?

SASSA has not published a detailed technical explanation, but the gap, nearly 7,800 complaints against almost none for fingerprint checks in one reported sample, suggests facial recognition is more sensitive to lighting, camera quality and image conditions common on lower-end smartphones.

Q3. Is the R4.8 billion underspend connected to the review programme?

Not directly, but critics argue it undermines SASSA’s fiscal case for tightening reviews on beneficiaries while billions in already-allocated grant funding went unspent in the same period.

Q4. Were the “ghost beneficiary” fraud claims accurate?

Mostly not. SASSA says the majority of roughly 74,000 flagged deceased-beneficiary accounts were reporting-timing mismatches rather than genuine payments made after death.

Q5. Does criticism of the review process mean fraud isn’t a real problem?

No. Confirmed fraud cases, including officials dismissed and beneficiaries prosecuted, are real and documented. The debate is over whether the current process fairly distinguishes fraud from ordinary administrative and access barriers.

Twenty years in, SASSA is defending a review programme it did not fully choose, against critics who are not actually arguing the programme shouldn’t exist. That is a narrower disagreement than the two opinion pieces might suggest at first read. The real argument is over execution: whether a facial scan that fails nearly half as often as it should get sent back into a queue, whether a beneficiary 40 kilometres from an office gets a fair chance to respond, whether R4.8 billion sitting unspent belongs anywhere near a conversation about tightening belts on pensioners.

SASSA’s own numbers, and its critics’ own numbers, mostly agree on the facts. What they disagree on is whether the current process treats the person on the other end of a review notice with the dignity SASSA says its entire twenty-year mission has been built around. That is not a question either opinion piece can fully answer from a newsroom. It gets answered, one review at a time, in queues this publication will keep covering as the 350,000-review target for this financial year plays out.