It took a judicial investigator, a unit of the Hawks - South Africa's Directorate for Priority Crime Investigation - and a specialised tribunal in Pretoria to set in motion a process that most international mechanisms would have kept in grey territory for years. On 2 March 2026, Director-General of the National Department of Health Sandile Buthelezi was arrested alongside two senior officials - Deputy Director-General Malixole Mahlathi and Chief Financial Officer Phineas Mamogale - and charged with fraud and theft involving funds allocated to South Africa by the Global Fund to Fight AIDS, Tuberculosis and Malaria. The prosecution was precise, substantiated, and - crucially - criminal: according to the charges, the trio allegedly diverted more than one million rand from a Global Fund project originally earmarked for Covid-19 psychosocial support to finance an internal disciplinary inquiry against a whistleblower.

What could have remained, yet another administrative scandal became a fully-fledged criminal case. It was an internal judicial investigator who referred the three accused to the Hawks, triggering a formal investigation. Arrests followed, then a specialised court, bail hearings, suspensions issued on the directive of President Cyril Ramaphosa, and the appointment of an acting Director-General. The national machinery worked. That is rare enough to warrant close attention - and to draw from it the lessons that bear directly on how the Global Fund currently conceives of accountability.

The Global Fund and its own mirrors

Since its creation in 2002, the Global Fund has built an impressive internal control architecture: an Office of the Inspector General (OIG) with a robust mandate, audit mechanisms for principal recipients, Local Fund Agents (LFAs), and periodic fiduciary risk assessments. This edifice is not without merit. It has exposed significant misappropriation - from the Philippines to Mali, from Tanzania to Mauritania - and secured repayments. It has helped establish the Fund's credibility with donors anxious to ensure their contributions reach target populations.

But this model rests on an implicit premise: national systems are insufficient to guarantee the integrity of funds. That premise has a history. It was forged in the 2000s on real cases of serious failure - unsanctioned fraud, misappropriation covered up by management hierarchies, absence of meaningful judicial recourse. These episodes gave rise to an institutional culture in which international surveillance is treated as the last line of defence against impunity.

The problem is that this culture tends to harden into dogma. In a few countries, national control systems - audit institutions, anti-corruption bodies, specialised prosecutors - have evolved considerably. Ignoring that evolution, or systematically underestimating it, perpetuates an asymmetric relationship that carries real costs: transaction costs for recipients, duplicated audits, and the symbolic undermining of the very national institutions the Fund purports to strengthen. Siloed accountability is not a long-term strategy - it is a defensive posture.

What Pretoria did that Geneva could not

The Buthelezi case illustrates with disarming clarity the limits of exclusively international accountability. What would have happened had the irregularities been detected by the Global Fund's OIG alone? Most likely: an investigation report, exchanges with the South African government to obtain repayment, possibly a temporary suspension of disbursements, and - in the best case - internal disciplinary measures. The OIG's tools are powerful in their own register: they document, expose, recommend. But they do not indict. They do not issue arrest warrants. They do not suspend officials under presidential directive.

It was the Hawks who arrested Buthelezi. It was the Specialised Commercial Crimes Court in Pretoria that set bail conditions. It was Health Minister Aaron Motsoaledi who ultimately suspended the three officials, following consultation with the Presidency. The chain of consequences - criminal, administrative, political - was only possible because it unfolded within the sovereign framework of South African law, with its own actors, procedures, and legitimacy. An international organisation, however well-resourced, cannot produce this kind of effect. It can accompany, support, and document it. It cannot replace it.

This point deserves precision: the South African case is not simply a story of detected corruption. It is a story of national accountability functioning - where an internal whistleblower triggered a formal criminal inquiry leading to high-level arrests. It is this movement - from internal to judicial - that the Global Fund cannot generate on its own. And that it has, until now, done too little to catalyse.

Genuinely shared accountability

The question is not whether to choose between the Global Fund's mechanisms and national systems. That is a false alternative. The real question is: how to articulate them strategically, considering the real capacities of each context, to maximise deterrent effects and ensure that bad actors face concrete consequences?

A more strategic approach would begin with a serious mapping of national oversight capacities - anti-corruption, audit, judiciary - in each recipient country, and adapt accordingly the level of Fund supervision. Where institutions function - as they demonstrably did in South Africa in this case - the OIG and LFAs can focus on supporting, informational, and strategic monitoring roles, rather than substitution. Where they remain deficient, international oversight stays indispensable but must be paired with an explicit investment in building the capacity of local actors.

The next step would be to formalise interfaces between Global Fund mechanisms and national judicial authorities. Too often, OIG findings remain confined to the international administrative circuit, never translated into national criminal proceedings. Clear protocols for transmitting information to competent prosecutorial offices - where those offices offer minimum guarantees of independence - would multiply the impact of internal investigations by giving them the follow-through that only national law can provide.

The Buthelezi case may not become a landmark. The three officials are presumed innocent until a court rules otherwise, and the hearing is not scheduled until June. The systemic context remains heavy - a Special Investigating Unit report from September 2025 had already revealed a vast corruption network at Tembisa Hospital potentially involving two billion rand. But what happened in Pretoria in March 2026 signals something the global health community cannot ignore: national institutions, when they function, can go further than any international mechanism ever could.

There is reason for cautious optimism. In recent funding cycles, the Global Fund has begun investing in Public Financial Management (PFM) - working with national treasuries, audit institutions, and oversight bodies to strengthen the systems through which its grants flow. This is a meaningful step: it signals a growing recognition that durable accountability must be rooted in national institutions, not merely supervised from above. The Buthelezi case - where it was a national forensic investigator, not an international auditor, who triggered the entire chain of consequences - is precisely the kind of evidence that validates this direction. It also underscores how long overdue it was. Deepening the Fund’s PFM investment, and systematically connecting it to the interface between oversight mechanisms and national judicial systems, would be the logical next step toward an accountability doctrine equal to the reality of a maturing global health architecture.

Christian Djoko Kamgain, PhD (Editor-in-Chief of GFO/OFM)