Voters' verdict looms as GNU's blame game takes center stage
Politics & Governance

Voters' verdict looms as GNU's blame game takes center stage

Coalition's collapse will hinge on who gets blamed for its failures

South Africans went to the polls in November hoping the governing arrangement might finally deliver jobs, working infrastructure and functioning public services. The question now is not simply whether the Government of National Unity will survive. It is who will be blamed when it ends, and how that argument will shape the lives of the people casting the next national vote.

Two things can be known with certainty about South African politics between now and 2029. The GNU, in its present form, will come to an end, and South Africans will elect a new national government. What remains genuinely uncertain is the timing. The political account of the GNU’s ending may well determine the outcome of that election, whenever it comes.

The negotiations following the 2026 local government elections have sharpened the question. The ANC, having suffered another substantial electoral defeat, is considering agreements with ActionSA, the MK Party, the EFF and other smaller parties that could exclude the DA from government in several of the country’s most important municipalities, most notably in the metros where the DA has become the largest party in all but three. There is even reason to believe these arrangements could provide the basis for a new governing majority nationally.

A collapse of the GNU does not necessarily mean an immediate national election, however. An alternative parliamentary majority, possibly made up of the ANC, ActionSA, the EFF and the MK Party, could replace the present arrangement without dissolving Parliament. Meanwhile, the Constitution offers routes to an early vote: a majority in the National Assembly can dissolve it once three years have passed since the 2024 election, and a presidential vacancy followed by Parliament’s failure to elect a successor within 30 days would also require dissolution. A new national election is therefore possible as early as 2027.

The stakes for ordinary voters are considerable. South Africans entered the GNU period with hopes that cooperation between the country’s largest parties might begin to reverse years of economic stagnation, unemployment, failing infrastructure and administrative decline. More than halfway through this parliamentary term, the evidence remains discouraging. Official unemployment stood at 33.7% in the third quarter of 2026, and the economy contracted by 0.2% in the second quarter. There is little reason, on present evidence, to assume the GNU will have delivered the sustained recovery its supporters hoped for by the time the next election arrives, particularly if it is brought forward.

The electoral ground has shifted dramatically. The ANC’s decline has continued well beyond the loss of its national majority in 2024. In November’s municipal elections it fell to below a third of the national vote share, while the DA rose to nearly 25% and became the largest party in five of South Africa’s eight metropolitan municipalities, including all three in Gauteng. The two parties are now within six points of each other, a gap that could close given the pace of the ANC’s collapse and the DA’s modest upward trajectory.

By contrast, the same developments that have improved the DA’s prospects have made it perhaps more vulnerable than ever. The DA is the only major party attracting substantial support from all four of the country’s principal racial population groups, while its competitors remain far more dependent on particular constituencies. The Freedom Front Plus has an obvious interest in winning white Afrikaans-speaking voters from the DA, as it managed in 2019, and the Patriotic Alliance seeks to displace the DA among coloured voters. The DA historically attracted only 4 to 5% of black voters’ support; this month it essentially doubled that, gaining 8% of votes cast by black voters. The ANC and ActionSA compete directly for those voters, while the MK Party and EFF have every reason to resist the DA’s further expansion, because its growth threatens the constituencies on which they depend.

The most damaging accusation against the DA would also be the simplest: it entered government promising change, South Africans saw too little improvement, and the GNU ended in recrimination. The ANC could use that account to escape responsibility for its own failures, while ActionSA and the PA could invoke it to justify making common cause with the ANC, without needing any formal electoral alliance.

The DA faces a demanding task whichever way events unfold. A party that walks away in unexplained anger risks being accused of choosing party advantage over national stability. A party that stays indefinitely despite repeated defeats on its key reform proposals will struggle to explain what its participation achieved. Even exclusion by the ANC and its new partners would leave the circumstances of its departure open to contest. It must demonstrate, with evidence, that it exhausted every reasonable opportunity to make the GNU and the 2026 negotiations work, offering a compelling account of what its ministers accomplished, where they met opposition, and which decisions were taken against their input, while acknowledging its own failures.

Voters, in the end, will have little patience for disputes conducted through coalition agreements and competing press statements. They will want to know why growth remained inadequate, why jobs were scarce, and why public services kept failing. If the DA can establish that it acted in good faith and the ANC chose to abandon collaboration, the party must ensure South Africans understand who made that choice and what it will do now. Such an account cannot be manufactured at the moment of collapse; it depends on decisions taken and explained throughout the remaining life of the government.

Two things will happen by 2029, even if the timing is unknown: the present GNU will end, and South Africans will elect a new national government. When they do, who will have won the argument over how the first of those two endings came about?

Q&A

Will the Government of National Unity survive until 2029?

No. The GNU in its present form will end before 2029, though the timing is uncertain. A collapse does not necessarily mean an immediate national election, since an alternative parliamentary majority could replace it without dissolving Parliament.

How could an early national election happen?

The Constitution allows a National Assembly majority to dissolve it once three years have passed since the 2024 election, and a presidential vacancy followed by Parliament failing to elect a successor within 30 days would also require dissolution. An election is therefore possible as early as 2027.

What changed in the November municipal elections?

The ANC fell below a third of the national vote share, while the DA rose to nearly 25% and became the largest party in five of South Africa's eight metropolitan municipalities, including all three in Gauteng, leaving the two parties within six points of each other.

What do voters want to know when the GNU ends?

Voters will want to know why growth remained inadequate, why jobs were scarce and why public services kept failing, rather than following disputes conducted through coalition agreements and competing press statements.

Related articles

  1. 1 Politics & Governance Odds of ANC Left-Wing Coalition Deal Rise to 75%, Says FCPC
  2. 2 Politics & Governance Big week for SA households: coalition talks and rates call
  3. 3 Politics & Governance Neighbors of ANC ex-presidents live the political shift
  4. 4 Politics & Governance ANC blames failing services as election support crumbles
  5. 5 Politics & Governance ANC's 30% local vote share rattles investors