Social protection remains one of the most visible, impactful, and fundamentally necessary pillars of South Africa’s democratic settlement. The post-apartheid state was established on a commitment to human dignity, equality, and the progressive realisation of socioeconomic rights. Chief among these is the right of access to social security, including appropriate social assistance for those unable to support themselves and their dependants, which is enshrined and protected in section 27(1)(c) of the Constitution of the Republic of South Africa. Three decades into democracy, this rights-based framework underpins a vast administrative system that serves approximately 26.5-million grant beneficiaries nationwide.

Poverty, inequality, and high unemployment make social protection essential in South Africa. According to Statistics South Africa’s 2025 Poverty Trends Report, about two-thirds of the population live below the upper poverty line of R2 635 per month, and more than a third survive on less than R1 300 per month. At the same time, around 12.4-million working-age adults, accounting for about 31.4 percent, lack regular work or income. The country’s Gini coefficient, estimated at approximately 0.65, confirms that South Africa has one of the most extreme levels of income inequality globally.

South Africa's social protection system rests on three pillars, comprising social assistance (non-contributory cash transfers), statutory social insurance funds, and voluntary social protection arrangements. Of these, social assistance, administered through the South African Social Security Agency (SASSA) under the Department of Social Development (DSD), represents the most significant mechanism for poverty alleviation. The Old Age Grant, Child Support Grant, Disability Grant, and other categories reach approximately 19.3-million beneficiaries, with the Child Support Grant alone covering roughly 13-million children.

Poverty, extreme inequality, and structural unemployment make comprehensive social protection an absolute necessity for survival in this challenging context. South Africa’s social assistance framework is the primary mechanism for poverty alleviation and redistribution. Programmes such as the Old Age Grant and the Child Support Grant have been recognised for their effectiveness in reducing extreme destitution and supporting household resilience. As highlighted in a comprehensive review by the World Bank, South Africa’s system of non-contributory cash transfers is generally well-targeted and provides substantial benefits to the poorest households. However, the system has historically provided almost no direct income support for unemployed, working-age adults who did not have a disability. It was only the onset of a global pandemic that forced a fundamental reckoning with this omission.

How the Social Relief of Distress Grant changed the conversation

The introduction of the Social Relief of Distress (SRD) grant in May 2020 marked a watershed moment in the history of South African social policy. Initially conceived as a temporary emergency measure to mitigate the devastating economic shock of the Covid-19 lockdowns, the SRD grant provided a modest monthly cash transfer to unemployed adults who were not receiving any other form of social assistance or unemployment insurance. Over the subsequent years, it evolved from an emergency intervention into a recurring programme. By the time of the 2026 Budget Review, the grant was supporting approximately 8-million monthly beneficiaries, with the Minister of Finance confirming a R36.4-billion allocation to extend the programme through to March 2027. The SRD grant fundamentally altered the national conversation regarding social protection in two distinct ways.

First, it breached the longstanding policy reluctance to provide unconditional cash transfers to able-bodied, working-age adults, demonstrating both the profound need for such support and its macroeconomic benefits. By injecting cash directly into the poorest communities, the grant stimulated local demand, supported informal traders, and prevented-millions from slipping into food poverty.

Second, and more importantly for the future of digital governance, the SRD grant represented South Africa’s first large-scale, predominantly digital social assistance programme. Forced by pandemic restrictions to avoid physical queues, SASSA had to design and launch an end-to-end digital system within weeks. Applications were exclusively accepted through digital channels, such as a website and a zero-rated mobile messaging application. Eligibility was determined through automated processes that cross-referenced applicant identification numbers against multiple government databases, including the South African Revenue Service (SARS), the Unemployment Insurance Fund (UIF), and the National Student Financial Aid Scheme (NSFAS). Payments were subsequently routed through commercial bank accounts, mobile money services, or retail cash-out points.

This digital-first approach provided a glimpse into the future of state capacity. It proved that the South African government could process-millions of applications, verify identities across disparate databases, and disburse funds digitally without relying on legacy, paper-based administrative systems. However, the very same digital systems that enabled its massive scale began to reveal deep-seated vulnerabilities. Its deployment soon exposed the severe risks of digitising social assistance without adequate data governance, inclusive design, and robust mechanisms for administrative justice.

The flaws the SRD grant experience exposed

As the SRD grant transitioned from a temporary measure into a medium-term programme, the government introduced increasingly stringent eligibility checks to manage the fiscal burden. These changes abruptly shifted the nature of the digital systems from mechanisms of inclusion to instruments of exclusion. The most critical operational failure to emerge was an extraordinarily high rate of erroneous exclusion. These are instances where individuals who were legitimately destitute were rejected by the automated system.

A central driver of these exclusions was the implementation of a rigid means test coupled with automated bank verification. To qualify for the grant, an applicant’s monthly income could not exceed a specific threshold (pegged initially at the food poverty line). To enforce this, SASSA utilised automated systems to scan applicants' bank accounts. As detailed in a policy brief by the Institute for Economic Justice (IEJ), the automated verification system lacked the nuance needed to distinguish genuine income from other types of monetary inflows. If an applicant received a small, sporadic cash transfer from a relative to buy electricity, or accepted a child maintenance payment on behalf of a dependent, the system flagged this inflow as income.

Research found that approximately 90 percent of all exclusions from the SRD Grant are erroneous, meaning the excluded individuals were in fact eligible. The primary driver is the bank income verification process, which cannot distinguish between types of monetary inflows, treating all deposits as "means" regardless of whether they constitute income, loans, maintenance payments, or intra-household transfers. The research found that 80 percent of all exclusions are due to bank income checks and that 76 percent of applications rejected on this basis were actually eligible.

This crude application of data-matching disproportionately penalised women, who frequently receive maintenance payments, and the poorest individuals who rely on informal, intra-household transfers to survive. The situation culminated in civil society organisations bringing a major legal challenge against the government. In a landmark 2025 judgment in Institute for Economic Justice v Minister of Social Development, the High Court declared several of these verification procedures and database checks unconstitutional. The Court observed that the automated system unlawfully double-counted intra-household payments and completely failed to account for income fluctuations, thereby denying legally entitled citizens their constitutional right to social assistance.

Beyond bank verification, the SRD experience laid bare the poor state of data quality and interoperability across the South African public sector. Because there is no comprehensive, unified social registry, SASSA had to rely on point-to-point data exchanges with other departments. Many of these databases were outdated. An applicant might be rejected because an obsolete UIF database incorrectly reflected them as employed, or a NSFAS database inaccurately flagged them as receiving a student stipend.

Furthermore, the digital-only application model created an immediate barrier for the most vulnerable. While applying via a mobile phone is convenient for the connected urban youth, it represents a significant hurdle for those lacking digital literacy, affordable data, or access to smart devices. By replacing physical service centres entirely with digital portals, the state inadvertently filtered out a segment of the population it was targeting.

When the automated system inevitably made mistakes, the avenues for redress were severely compromised. The Promotion of Administrative Justice Act (PAJA) requires that administrative action be lawful, reasonable, and procedurally fair, which includes the right to be given written reasons for a negative decision and a fair opportunity to challenge it. However, the SRD appeals process was entirely digital and highly restrictive. Applicants received generic rejection codes rather than detailed reasons, and the automated appeals portal did not allow them to submit new evidence (such as a bank statement proving a deposit was a loan, not a salary). Consequently, the appeals system functioned more as a repetitive algorithmic loop than a platform for meaningful administrative justice.

The volume of appeals exceeded 10-million in the last financial year, with a 98 percent rejection rate, suggesting systemic barriers to lodging successful appeals. Applicants who remain dissatisfied after the Independent Tribunal have only the High Court as recourse, which is beyond the means of virtually all SRD applicants. Effective redress mechanisms are essential for maintaining trust in automated public services, yet the current system appears designed more for administrative efficiency than for justice.

Why payments matter as much as policy

In discussions of social protection, the focus overwhelmingly falls on policy parameters such as who qualifies, what the threshold should be, and how much the grant is worth. The actual nuts and bolts of the system, insofar as how money flows, physically or digitally, into a citizen's hands, are often treated as a technical issue. The SRD grant experience proved that payment design is profoundly political. It shapes the cost, dignity, timing, choice, and level of dependence experienced by the beneficiary.

Historically, the South African grant system relied heavily on the South African Post Office and its banking subsidiary, Postbank. However, as Postbank faced severe leadership crises, cryptographic security breaches, and systemic operational failures, the payment ecosystem became highly fragmented. To ensure the SRD grant reached-millions of people, the government had to pivot, leaning on the private sector. Beneficiaries were encouraged to route their payments into commercial bank accounts, mobile money wallets, or to collect cash directly from large retail supermarket chains.

While this diversification prevented a collapse of the disbursement system, it introduced new complexities and dependencies. When commercial banks or mobile network operators control the payment rails, they naturally apply their own fee structures. For a grant recipient receiving a marginal sum, even a minor withdrawal fee or account management charge represents a significant erosion of the benefit’s value. Furthermore, relying on retail supermarkets as cash-out points subtly alters the power dynamic. Retailers benefit from guaranteed foot traffic, with beneficiaries spending their grant money in-store rather than freely withdrawing the full cash amount.

The reliance on third-party intermediaries also raised critical questions regarding the ownership and monetisation of beneficiary data. Private platforms that facilitate grant payments or zero-rated access often collect vast amounts of location data, transaction history, and device information. Without stringent oversight, the poorest citizens effectively pay for their access to state support by surrendering their digital privacy to commercial entities.

If the payment rail is expensive, fragile, or captured by rent-seeking intermediaries, the social policy is itself undermined. Transforming social assistance, therefore, requires building a payment infrastructure that treats the secure, low-cost transfer of public funds as a public good rather than a commercial opportunity.

Building the digital spine

Recognising the severe limitations of siloed databases and fragmented payment systems, the South African government has adopted a more systemic technological vision. This vision is heavily influenced by the global concept of Digital Public Infrastructure (DPI). Unlike traditional IT procurement, which builds standalone software for specific departments, DPI refers to foundational, shared digital building blocks, including digital identity, data exchange platforms, and payment utilities, that can be utilised by any government service or private sector actor to deliver services safely and efficiently.

This approach has been formalised in South Africa’s Roadmap for the Digital Transformation of Government, published in 2025. The roadmap outlines a phased strategy to modernise public service delivery. Phase 1 (2025–2027) explicitly prioritises social protection, aiming to build a functional digital identity system, an interoperable data exchange, and unified digital channels for citizens to interact with the state.

Simultaneously, the South African Reserve Bank (SARB) is advancing its Payments Ecosystem Modernisation (PEM) programme. The PEM programme includes the rollout of rapid payment systems like PayShap, but its most transformative element for the state is the proposed establishment of a Public Payments Utility. This utility aims to provide open, shared digital payment infrastructure that operates outside the exclusive control of major commercial banks.

PayShap's relevance to social grant payments lies in its potential to reduce costs, increase speed, and improve financial inclusion. Beneficiaries without traditional bank accounts could potentially receive payments using only their mobile number as an identifier, reducing dependency on formal banking relationships. SARB has estimated that the RPP could add approximately 0.25 percent to GDP by reducing the indirect costs of cash transactions and improving payment system efficiency.

The Public Payments Utility (PPU), also referred to as the National Payments Utility, represents a more ambitious component of the PEM Programme. The PPU is being established through SARB's 50 percent acquisition of PayInc (formerly BankservAfrica), the country's primary payment clearing house. It is intended to provide open digital payments infrastructure to a broader range of participants, including non-bank financial services providers. The PPU's proposed functions include operating the core payment-clearing infrastructure, setting scheme rules and standards, enabling interoperability between payment providers, and facilitating the entry of new participants into the payment ecosystem.

The relationship between the Rapid Payments Programme and the Public Payments Utility is complementary but distinct. PayShap operates at the retail payment layer, providing instant, low-cost transactions for consumers and businesses. The PPU operates at the infrastructure layer, providing the clearing and settlement mechanisms that underpin payment systems. For grant payments, the logical flow would be that SASSA initiates a bulk payment instruction, the PPU routes the instruction to relevant payment providers, and PayShap delivers funds instantly to beneficiaries' accounts or mobile wallets. Beneficiaries could receive payments using their mobile number as a proxy identifier, eliminating the need to remember bank account details.

If implemented successfully, this DPI stack could resolve many of the structural problems exposed by the SRD grant. A verified, biometric digital identity could drastically reduce application fraud and eliminate the need for citizens to continuously present physical documents at government offices. An interoperable data exchange would allow SASSA to securely and instantly query real-time data from SARS or the Department of Home Affairs, rather than relying on outdated batch files. This would drastically reduce the exclusion errors caused by obsolete employment records.

Furthermore, integrating social grant disbursements with the SARB’s new rapid payment rails could improve the beneficiary experience. Instead of waiting days for batch electronic funds transfers (EFTs) to clear, or paying high fees to withdraw cash from ATMs, a beneficiary could receive their grant instantly via a simple proxy identifier, such as their mobile phone number. A public payments utility would ensure that these government-to-person (G2P) transfers are processed at the lowest possible cost, maximising the amount of the grant that actually reaches the beneficiary’s pocket.

The governance questions that cannot be ignored

However, technology alone cannot resolve issues of social justice. The same digital infrastructure that promises efficiency and inclusion carries inherent risks of surveillance, exclusion, and unchecked administrative power. The lesson of the SRD grant is that when digital reform is implemented without robust governance, it inevitably hardens exclusion.

The first critical risk is the unchecked expansion of automated decision-making. When a citizen’s livelihood is determined by an algorithm querying an interoperable data exchange, the transparency of that process is significant. The Protection of Personal Information Act (POPIA) provides foundational safeguards for data privacy, mandating that information be collected for specific purposes and that data subjects be informed. POPIA, however, struggles to fully address the complexities of state-level algorithmic profiling. If the government integrates multiple databases to create a 360-degree view of a citizen to determine welfare eligibility, it risks building an invisible, unaccountable surveillance apparatus.

There is a documented risk that data matching can be exploited to artificially constrain beneficiary numbers. As evidence in the IEJ court case demonstrated, strict algorithmic parameters and crude bank verifications were utilised partly because National Treasury had imposed strict budget caps on the SRD grant. When fiscal consolidation dictates policy, automated systems provide a convenient, obscure mechanism for quietly turning people away without human intervention.

Finally, the drive toward digital efficiency must not ignore the reality of the digital divide. Imposing mandatory biometric verification via smartphones, or requiring digital-only identities, risks entirely disenfranchising the rural elderly, the undocumented, and the severely impoverished. A system that works perfectly for a connected urban citizen but excludes a vulnerable rural citizen is a failure of public policy, regardless of its technological sophistication.

Designing for dignity

To ensure that the next phase of digital reform widens access rather than hardening exclusion, South Africa must deliberately design its digital public infrastructure around the principles of dignity, rights, and administrative justice. This requires moving beyond a purely technocratic view of digitisation and embracing a rights-based reform agenda.

First, the state must institutionalise "assisted digital" and non-digital pathways as permanent features of the service landscape. Digital self-service portals should be viewed as tools to reduce the burden on physical offices, thereby freeing up human caseworkers to assist those who cannot navigate the digital system. A citizen’s inability to access the internet must never result in the forfeiture of their constitutional right to social security.

Second, the logic of administrative justice must be hardcoded into automated systems. In compliance with PAJA, any algorithmic decision that results in the denial of a social grant must generate a clear, comprehensible written reason for the applicant. More importantly, the appeals process must include a mechanism for meaningful human review. Citizens must be permitted to upload supplementary documentation or present contextual evidence to human adjudicators when automated data-matching produces an unjust outcome.

Third, the governance of data exchange must be strengthened through explicit legislative frameworks. The current reliance on fragile Memoranda of Understanding between disparate government departments is insufficient. Clear statutory duties must be established to govern how data is shared, updated, and contested, ensuring that the burden of correcting state data errors does not fall solely on impoverished citizens.

Finally, the governance of the payment ecosystem requires urgent attention. As the SARB’s PEM programme advances, the state must mandate transparent pricing structures and strict data protection clauses for any private entity involved in processing social grants. A public payment utility should serve to guarantee a basic, fee-free transfer mechanism for welfare beneficiaries, ensuring that commercial intermediaries cannot extract rent from public assistance.

The road ahead

The expanding need for income support, as highlighted by the SRD grant, and the rollout of sophisticated digital public infrastructure, such as the MyMzansi roadmap, present a unique opportunity to rewrite the social contract. The lessons drawn from the last several years are clear.

Technology has the power to dismantle bureaucratic silos, route cash instantly to those in need, and provide the state with the tools necessary to administer social assistance grants on a massive scale. However, when digital systems are deployed primarily for cost-containment, and operated without stringent safeguards, they function as invisible walls, systematically excluding the most vulnerable citizens.

As South Africa builds its digital identity frameworks, data exchange platforms, and payment utilities, the defining challenge will not be technical feasibility, but political governance. The ultimate measure of the country’s digital transformation will be whether the resulting infrastructure disciplines and excludes the poor, or whether it fulfils the constitutional promise of an administration that is lawful, reasonable, procedurally fair, and dignified.

Written by Mark Burke, a researcher and advisor with expertise in digital governance, and a focus on public-sector digital transformation. His research interests are digital identity, privacy, and citizenship in the digitalisation of public services.