Something is shifting in South African politics, and the data is starting to tell the story before the pundits catch up. For the first time since 1994, a party other than the ANC leads in national opinion polling. The Government of National Unity, less than two years old, is showing cracks that look structural rather than cosmetic. And at the municipal level, the upcoming local government elections, scheduled between November 2026 and February 2027, are forcing every party to choose between coalition loyalty and electoral survival. Nowhere is this tension more visible than in Cape Town.

South Africa's Government of National Unity was born out of electoral arithmetic, not shared vision. When the ANC lost its parliamentary majority in May 2024, dropping to 40.2% from 57.5% five years earlier, it assembled ten parties into a coalition centred on a broad Statement of Intent. The DA, IFP, Patriotic Alliance, Good, UDM, Freedom Front Plus, Al Jama-ah, PAC, and Rise Mzansi all signed on. The arrangement kept Cyril Ramaphosa in the presidency and gave the DA six cabinet portfolios, including Agriculture and Basic Education.

The coalition has faced significant tests. In early 2025, the national budget had to be tabled three times after the ANC and DA deadlocked over a proposed VAT increase from 15% to 17%. The Johannesburg Stock Exchange shed roughly R1 trillion in value during the standoff. The underlying problem, as Oxford Economics diagnosed it, is that the GNU was assembled without a dispute-resolution mechanism, and that structural flaw has never been addressed.

Ramaphosa's personal engagement has been inconsistent. At one point, 143 days passed without a formal meeting between the president and GNU party leaders, prompting a retreat at the Cradle of Humankind where Ramaphosa reportedly urged partners to maintain unity. A formal Leaders Forum was subsequently established and met in October 2025, producing a joint statement affirming the GNU's stability. The statement committed the parties to closer coordination but offered few specifics on policy direction.

The immediate pressure now is electoral. With local government elections on the horizon, GNU partners must campaign against each other in municipalities while governing together nationally. The ANC and DA will compete for metros like Johannesburg and Tshwane, and the Patriotic Alliance has already named mayoral candidates for both Johannesburg and Cape Town. The inherent tension of being allies in Pretoria and rivals on the ground will only increase as election day approaches.

One of the most significant data points in South African politics right now is a number that would have been difficult to imagine five years ago. In April 2025, an Institute of Race Relations opinion poll placed the DA at 30.3% national support, edging past the ANC at 29.7%. It was the first time any party had polled ahead of the ANC in the democratic era.

The context matters. The poll's fieldwork ran from 27 March to 3 April 2025, during the peak of the VAT standoff. The DA had positioned itself as the party opposing a tax hike on struggling households, and voters responded. Among black registered voters, DA support rose from 5% to 18%, a shift the IRR described as evidence of a move from identity-based to bread-and-butter politics.

Whether that shift is durable is the open question. Writing in March 2026, the Daily Friend's analysis argued that the DA's surge was structurally tied to a single, concrete economic issue, and that as the VAT debate faded from headlines, identity-based political framing reasserted itself. The implication is that the DA needs to keep the contest focused on economic terrain (jobs, prices, services) or risk seeing its polling gains recede before a single ballot is cast.

Into this national picture steps Geordin Hill-Lewis, Cape Town's 37-year-old mayor, who on 27 February formally entered the race to lead the DA. He is effectively running unopposed: the only other nominee, Sibusiso Dyonase, is a municipal councillor from Sedibeng with limited national profile. The vote takes place at the DA's federal congress on 11-12 April in Johannesburg.

Hill-Lewis's candidacy follows the exit of John Steenhuisen, who confirmed on 4 February that he would not seek a third term. Steenhuisen framed his departure as a completed mission, having taken the DA into national government for the first time. Analysts noted, however, that the move was partly shaped by internal dynamics. The party had not grown significantly under his leadership: 22.2% in 2014, 20.8% in 2019, 21.8% in 2024. Movement within the margin of error across a decade is not a compelling growth story.

Hill-Lewis plans to hold both roles, DA leader and Cape Town mayor, at least through the local elections. It is a model Helen Zille used from 2009 to 2011, though the context is different now. The DA governs Cape Town with a stable majority, and the city consistently ranks as one of the best-run metros in the country. Under Hill-Lewis, Cape Town was named Best City in the World by The Telegraph Travel Awards. The mayoralty also faces criticism, however: housing affordability is a growing pressure point, and the city's murder rate remains among the highest in the country.

The broader question is what Hill-Lewis's leadership means for the DA's relationship with the ANC inside the GNU, and for Cape Town itself. A party leader who is also the country's most prominent opposition mayor creates an unusual dynamic. Hill-Lewis has already signalled that the GNU needs to move faster on reforms, warning in January that progress on unemployment and fiscal strain is too slow. His leadership of the DA during a contested election campaign could put additional strain on the coalition's cohesion.

The Independent Electoral Commission confirmed in March that voter registration weekends will be held on 20-21 June 2026, with the elections themselves falling between November 2026 and February 2027. As of December 2025, 27.67 million South Africans were registered to vote. The largest voting bloc is citizens aged 30 to 39, and 55% of registered voters are female. Some 508 political parties have registered to participate.

The fragmentation is notable. With the ANC, DA, MK, EFF, Patriotic Alliance, ActionSA (now merged with Forum 4 Service Delivery), and Floyd Shivambu's new Afrika Mayibuye Movement all contesting, the electoral map will be more divided than ever. Currently, only three of South Africa's eight metropolitan municipalities (Cape Town, Buffalo City, and Mangaung) are not governed by coalitions. After these elections, the number could be zero.

For Cape Town specifically, the election is less about whether the DA retains control (it almost certainly will) and more about margins and mandate. The Patriotic Alliance has nominated Liam Jacobs, a recently sworn-in councillor, as its Cape Town mayoral candidate. The ANC will work to hold onto its traditional support base in the Cape Flats. The real contest may be about turnout: if the DA can demonstrate broadened support across Khayelitsha, Mitchells Plain, and Gugulethu, it strengthens Hill-Lewis's position as a national leader claiming to run a city that works for everyone.

Cape Town's political identity is built on a paradox that the data makes impossibly stark. The city wins international awards for liveability and tourism. It is Africa's tech hub, a magnet for digital nomads, home to a startup ecosystem that generates real venture capital. Under DA governance, the city has maintained relatively competent service delivery compared to Johannesburg or eThekwini, where water crises and infrastructure collapse are now routine.

And yet: four of the top five police precincts with the highest murder rates in the country are in Cape Town.

In Q2 of the 2025/26 financial year (July to September 2025), the Western Cape recorded 1,160 murders, a 9.1% increase year-on-year. Cape Town accounted for 967 of those, or 83.4%. Firearms were used in 60.6% of all murder cases. Mfuleni alone recorded 84 murders in Q2, the highest of any single precinct in the country, driven by gangsterism, extortion, and taxi violence.

There is a counterpoint, however, and it matters. In the areas where the Western Cape Government has deployed its Law Enforcement Advancement Plan (LEAP) officers (Delft, Gugulethu, Khayelitsha, Mitchells Plain, Nyanga, and Philippi East), murder rates fell 9.4% in Q2 and 3.7% in Q3. In the reaction unit areas of Elsies River, Manenberg, and Steenberg, the combined reduction was 14.3%, with Steenberg recording a 56.3% drop. The provincial government's argument is clear: locally managed, data-led policing works, but it needs national funding and the devolution of policing powers to scale.

This is the political fault line that the local elections will be fought on in Cape Town. The DA can credibly claim that its governance model delivers results where it is funded and empowered. But it governs a city where the overall murder trajectory is up, where housing costs are displacing long-standing communities from neighbourhoods like Salt River and Bo-Kaap, and where the distance between the Atlantic Seaboard and the Cape Flats remains one of the most visible expressions of inequality on earth.

The macroeconomic backdrop is cautiously positive. South Africa exited the FATF grey list in 2025, a milestone that had been years in the making. Inflation eased to 3.5% by early 2026, the lowest level in roughly two decades. The Reserve Bank has been cutting interest rates. S&P Global upgraded the country's credit rating. Four consecutive quarters of GDP growth, modest at between 0.4% and 0.6% per quarter, represent the most sustained expansion in several years.

The headline growth rates, however, sit alongside deeper structural problems. Unemployment fell to 31.4% in Q4 2025, the lowest since Q3 2020, yet youth unemployment (ages 15 to 34) still exceeds 46%. GDP per capita had been declining since 2014 when adjusted for population growth, and only turned positive again in 2025. The OECD projects 2026 growth at 1.4%, well below the 3%+ rate economists say is needed to meaningfully reduce unemployment. Deloitte notes that political uncertainty, tariffs, and low investment levels will constrain growth even as reform momentum builds.

The fiscal picture has improved marginally. South Africa has achieved two consecutive years of primary budget surplus, meaning the government takes in more revenue than it spends when debt-servicing costs are excluded. Debt-servicing costs remain substantial, though, and the revenue gains are partly due to commodity price windfalls (gold, platinum group metals) rather than structural improvements in tax collection.

For Cape Town, the economic picture is somewhat brighter. The city has positioned itself as Africa's tech capital, with a growing startup ecosystem, strong tourism recovery, and relatively functional infrastructure. But the cost of living pressure is real. Gentrification is pushing legacy communities out of central neighbourhoods. And the broader Western Cape economy, while outperforming many provinces, still operates within the constraints of a national system that produces 31% unemployment and 46% youth joblessness.

South Africa in March 2026 is a country where the macro indicators are mostly moving in the right direction, albeit slowly, while the political structures designed to sustain that progress are under increasing strain. The GNU is not collapsing, but it is not yet functioning as a coherent governing coalition. It persists in part because no party wants to bear the blame for ending it, which is a different thing from shared purpose.

The local elections will be the first genuine electoral test of this arrangement. They will indicate whether the ANC's decline, documented in the polling, translates into actual losses at the municipal level. They will also show whether the DA's growth among black voters was a passing reaction to the VAT debate or the beginning of a durable political realignment. And they will determine whether Cape Town remains the DA's strongest proof of concept for opposition governance, or whether the murder statistics, housing costs, and inequality data define the narrative instead.

For Cape Town specifically, the next twelve months will test whether the city's political model (competent DA governance layered over deep structural inequality, with a violent crime problem it cannot solve without national cooperation) can hold together under the pressure of an election campaign where every party has an incentive to highlight the contradictions.

Hill-Lewis, if he becomes DA leader as expected, will carry the weight of those contradictions into the national arena. He will be simultaneously the face of municipal competence and the mayor of a city where 967 people were murdered in a single quarter. That is not a political problem to be managed through messaging. It is the condition of South African governance itself: real progress and real dysfunction, present at the same time.

The numbers will keep moving. The politics will keep shifting. What won't change is the underlying question that every election in this country asks, and that none has yet convincingly answered: can South Africa build a politics that is about material progress for the majority, not just identity or legacy or inherited loyalty? The April 2025 polling suggested, for a brief moment, that it might. Whether that moment extends into something lasting is the story of the next year.