Not a Zero-Sum-Game: Poll Hints That Both DA and ANC May Gain from the GNU

Staff Writer

February 2, 2026

4 min read

New polling suggests the unity government is being judged pragmatically, with both major partners drawing credit for co-operation.
Not a Zero-Sum-Game: Poll Hints That Both DA and ANC May Gain from the GNU
Image by Jürgen Bierlein from Pixabay

South African voters are close to evenly divided on how the African National Congress (ANC) is performing inside the Government of National Unity (GNU), while viewing the Democratic Alliance (DA) more positively.

These are the results of a Social Research Foundation* poll conducted in late 2025.

Asked whether the ANC is performing very well, quite well, quite poorly, or very poorly in the GNU, 41.3% of respondents say it is doing very well or quite well. A slightly larger 44.9% say it is doing quite poorly or very poorly. The remainder are undecided or declined to answer. The result amounts to a near even split, with voters divided down the middle on whether the ANC’s participation in the coalition is working.

That balance is politically significant. Coalition governments often generate decisive public reactions, either strong approval when governance improves or sharp disillusionment when cooperation falters. Instead, the ANC’s numbers sit in a narrow band between endorsement and rejection, suggesting a country still weighing the trade-offs of shared power rather than delivering a final verdict.

The DA’s assessment is more favourable. A combined 60.6% of respondents say the party is performing very well or quite well in the GNU, while 21.9% say it is performing quite poorly or very poorly. Even so, a sizeable undecided group indicates that many voters are still watching how the party exercises influence within government rather than rushing to judgement.

Taken together, the data suggests voters are not instinctively hostile to the GNU arrangement itself. Instead, they appear to be judging parties on visible performance and cooperation rather than ideology or historic loyalties.

Frans Cronje, speaking to The Common Sense, said, “This data bodes well for the future of the GNU and demonstrates that both parties may benefit from co-operation even as they remain rivals.”

Cronje added that the polling hints at a deeper dynamic that many commentators miss. The GNU is not necessarily a zero-sum contest in which one party must lose for the other to win. If the coalition measurably improves delivery, stabilises institutions, and reduces day-to-day governance failure, both major partners can gain at the same time. The ANC can begin to rebuild credibility by showing it can share power and still govern, while the DA can demonstrate it is capable of exercising influence nationally and improving state performance without collapsing government.

According to Cronje, “In that sense, the mutual advantages of cooperation could be substantial. A functioning GNU does not have to erase rivalry, and it does not require perfect ideological convergence. It simply requires enough shared interest in restoring basic competence that voters can feel the difference in their daily lives. If that happens, the coalition partners may find that the political rewards of stability and improved performance outweigh the short-term incentives to sabotage one another.”

*The Social Research Foundation’s Q4 2025 Market Survey was commissioned by the Foundation and conducted by Victory Research among 1 002 registered voters between 27 October and 14 November 2025 using telephonic CATI interviews. A single-frame random digit-dialling design was used, drawing from all possible South African mobile numbers to ensure that every number had an equal probability of selection, with national sim card penetration exceeding 250%, more than 90% of adults owning a phone, and mobile networks covering 99.8% of the population, giving universal practical coverage. The sample was fully weighted to match the national registered voter population across all key demographics, including language, age, race, gender, education, income, and urban or rural location. Turnout modelling assigned each respondent a probability of voting based on questions measuring their likelihood of participation, with the primary turnout model set at 52.8%. The poll carries a 4.0% margin of error at a 95% confidence level, with a design effect of 1.762.

Categories

Home

Opinions

Politics

Global

Economics

Family

Polls

Finance

Lifestyle

Sport

Culture

InstagramLinkedInXX
The Common Sense Logo