The Common Sense has set out three possible paths for South Africa through to 2034. Its most likely, or benchmark scenario, gives a 65% probability to continued coalition government, weak reform, and economic growth of roughly 1% to 2%. Businesses and communities increasingly provide for themselves where the state falls short, creating pockets of prosperity that prevent a wider collapse.

In its upside scenario, assigned a 20% probability, new African National Congress (ANC) leadership works effectively with the Democratic Alliance (DA), investment recovers, and growth reaches 4% to 5%. In the downside scenario, assigned 15%, the ANC aligns with parties to its left, economic policy becomes more interventionist, and the country enters a prolonged decline.

The benchmark deserves the closest attention, both because The Common Sense considers it the most likely outcome and because it rests on a striking claim: South Africans can adapt to a weakening state well enough to prevent national collapse.

I find that argument especially thought-provoking. South Africans can already see businesses, communities, and local institutions stepping in when the state fails to provide a service. Their ability to adapt is real. The question is how far it can carry the country as the pressures on them grow.

That is an insight worth taking seriously. It would be a mistake to assume that every failure of government translates immediately into failure throughout society. My concern is about the limits of this adaptation. Two pressures, in particular, could make the benchmark less stable than its assigned 65% probability suggests.

Unemployment

The first is the growing number of people without work. The benchmark envisages national growth of around 1% to 2% and unemployment remaining near 30%. Even if the unemployment rate stays broadly steady, that does not mean the number of unemployed people will do the same. As the labour force grows, more people can be without work while the percentage changes very little.

The Inclusive Society Institute’s 2026 paper, South Africa’s Population Dynamics, Economic Growth, and Employment Prospects, 2030-2045, illustrates the point. Under its 1% annual GDP growth scenario, it projects that the number of unemployed people will rise from about 7.7 million in 2025 to 9.2 million in 2045, even though the unemployment rate edges down from 32.4% to 31.6%.

These are projections based on stated assumptions, not a claim that the future can be known precisely. Their relevance to the benchmark is that a country may appear to be holding its unemployment rate steady while adding large numbers of people who cannot find work. That matters for the scenario’s expectation that political volatility will remain contained.

Prosperous areas can generate jobs, pay taxes, and support surrounding communities. But we should ask whether they can do enough of that to maintain social stability as the number of excluded people grows. Frustration may take the form of protest, crime, or stronger support for parties promising a different economic settlement. Any of these could affect the investment and confidence on which the enclaves themselves depend. The benchmark would be stronger if it explained why these pressures are likely to remain manageable.

The second pressure is the condition of the systems that connect prosperous areas to the rest of the country. Private security and backup power can make a substantial difference. So can capable municipalities and community organisations. But businesses still need roads, rail, ports, water networks, policing, and courts. Goods must move between places, contracts must be enforced, workers and customers must be able to travel.

There is room for greater private participation in some of these systems. That possibility belongs in the scenario analysis. But it also raises practical questions: which national functions could be provided privately, at what scale, and how quickly? What happens where private provision depends on a functioning public regulator, a shared network, or continued access to the wider economy?

The article suggests that enclaves become stronger as the central state becomes weaker. That may be true over a considerable range of state failure: people adapt because they have to. It is harder to assume that the relationship continues indefinitely. Beyond some point, deterioration of shared infrastructure and institutions becomes a constraint even for those best equipped to adapt. Identifying that point would help clarify how resilient the benchmark really is.

Further Question

There is a further question about the boundary between the benchmark and the downside pathway. The downside scenario focuses on an ANC leadership transition that moves the party leftward and into cooperation with more radical parties. Electoral incentives could provide another route to such a coalition.

Ipsos’s latest release, published on 4 September and based on interviews in June and July 2026, puts the ANC at 31%, the DA at 25%, the uMkhonto weSizwe Party (MKP) at 14%, and the Economic Freedom Fighters (EFF) at 11%. These are countrywide voting intentions among registered voters expressing a preference, not a forecast of a national election. But the ANC’s lead is only six percentage points. If the DA stayed at 25% and the ANC lost seven points to other parties, their positions would reverse. The ANC, MKP, and EFF together account for 56% in this poll, illustrating the potential coalition arithmetic.

Consider a future election in which the DA wins more votes than the ANC. The two parties might still command a majority together. But the ANC would face a choice: govern as the smaller partner in a DA-led arrangement, or seek to lead a coalition with parties to its left. The latter could be more attractive to leaders determined to retain control of government, regardless of whether it offers the better economic prospect.

There is also evidence of an opening for parties to the ANC’s left. In Ipsos’s September survey, 20% of ANC supporters named the EFF as their second choice, 9% the MKP, and 14% the DA. The EFF and MK together attracted 29%. Second choices do not predict defections, but they help identify where disappointed ANC voters might turn.

What might turn disappointment into defection? Consider an ANC government seen to be abandoning redress under American pressure. Solidariteit’s proposals to phase out corrective measures go further than the current American request for equity-equivalent arrangements for all American companies. But would that distinction survive an election campaign?

The EFF and MKP could portray either concession as surrender to Washington. Reform that South Africans accept as their own decision may have a different political effect from reform perceived as imposed from abroad.

A report released by the Thabo Mbeki Foundation released at the end of last month adds a warning about the depth of possible resistance. It argues that abandoning a determined programme to eradicate the legacy of colonialism and apartheid would provoke a social explosion worse than the American riots of the 1960s. We need not accept violence as inevitable to ask what happens if that resistance first finds expression at the ballot box. It is another reason to examine how the benchmark could give way to the downside pathway.

I am not suggesting that the ANC would necessarily make that choice. I am suggesting that an ANC-DA majority on paper cannot, by itself, tell us which government would be formed. The parties’ relative size and their incentives matter. If the ANC falls below the DA, the possibility of a leftwing coalition may deserve greater weight than the scenarios currently give it.

These observations build on the value of The Common Sense’s exercise. The three pathways provide a framework for asking the questions that will determine which future South Africa approaches. How many jobs would growth of 1% to 2% create relative to the expanding labour force? How much pressure would a growing number of unemployed people place on social stability? Which shared systems are essential to the continued success of the enclaves? Could externally driven changes to redress alter electoral loyalties? And how would an ANC that was no longer the largest party choose between possible coalition partners?

The benchmark may still prove to be the most likely pathway. Testing these questions would give us greater confidence in that judgement, or show where its probability should be reconsidered. Either result would add to the value of an already useful scenario-planning exercise.

Daryl Swanepoel is a South African political strategist who currently serves as the CEO of the Inclusive Society Institute. His political career began in the pre-1994 apartheid-era National Party, later culminating in his role as the final Secretary-General of the rebranded New National Party. Following this, Swanepoel joined the African National Congress, where he was employed directly at the party's national headquarters, Chief Albert Luthuli House, as a senior strategist and administrator. He also served as a Member of Parliament for the ANC. He serves as an advisor to Tianyuan University in China.