Pension Nationalisation Poll Exposes the Stabilising Power of Asset Ownership

A poll conducted in late 2025 by the Social Research Foundation (SRF)* found South Africans almost evenly divided on whether all pension funds should be nationalised and run by the state.

Among all voters, 42% agree strongly that pensions should be nationalised and run by the state, while 9% agree somewhat, taking total support to roughly half of all voters. Opposition is close behind, with 45% disagreeing strongly and 2% disagreeing somewhat, taking total opposition to 47%.

Those topline numbers look like a nation arguing with itself. But when the same question is viewed through the groups most likely to have private pensions, the argument ends.

Among South Africans earning between R40 000 and R80 000 per month, 2% agree strongly that pensions should be nationalised and run by the state and 8% agree somewhat, while 71% disagree strongly and 7% disagree somewhat. Among those earning above R80 000 per month, support collapses further. Only 1% agree strongly and 3% agree somewhat, while 90% disagree strongly and 6% disagree somewhat. In other words, the people most exposed to pension fund coercion reject it overwhelmingly.

This is the deeper lesson in the data. Asset ownership changes politics. When people build up pensions, savings, and a stake in formal financial markets, they become far less tolerant of policies that threaten long-term security. They stop responding to slogans about using savings for job creation and start asking whether their retirement will still be there when they need it.

This has practical relevance for South Africa, where the African National Congress has, from time to time, flagged in public discussion prescribed assets. Prescribed assets do not require nationalisation. It works by forcing pension funds and financial institutions to allocate a defined share of their portfolios into state-preferred instruments or projects. It is still compulsion. It still overrides normal investment choice. And it is economically negative because it substitutes confidence with coercion.

Compulsion raises perceived risk, pushes up the cost of capital, distorts investment allocation, and weakens incentives to save inside the formal system. It also encourages capital to seek safer jurisdictions or structures, leaving the domestic economy with less investable money, not more. If South Africa wants retirement capital to flow into productive investment, the route is improved policy that makes investment rational, not forced allocation that makes investment fearful.

The SRF figures therefore point to a strategy for stability. South Africa needs a broader middle class with pensions, savings, and property, a country of more asset owners who have something tangible to lose if policy turns predatory. The more people who own, the more the politics of the country shifts toward stability, predictable rules, and institutional competence. That is not a moral argument. It is a practical one. The data shows that asset ownership is one of the strongest anchors of long-term economic stability.

*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.