Every few years the same question returns to the South African investment debate, dressed in slightly different language. How do we direct more of the country’s substantial pool of retirement savings towards the infrastructure the economy so plainly needs?
The savings and investment industry holds several trillion rands on behalf of members, and the infrastructure financing gap runs to several trillion rands over the remainder of this decade. The temptation to connect the two by regulation is understandable.
The scale of the need, and the scale of the savings
Two numbers frame everything that follows. On the need side a joint World Bank and Development Bank of Southern Africa report published in January 2024 and still widely cited, estimated that South
Africa must spend between R4.8 trillion and R6.2 trillion between 2022 and 2030 on transport, water and sanitation, and education alone, equivalent to roughly 8.7% to 11.2% of GDP a year. Infrastructure South Africa has separately told Parliament that meeting the National Development Plan’s target of raising gross fixed capital formation to 30% of GDP by 2030 requires around R1.7 trillion of additional investment. The Minister of Finance has committed public infrastructure spending of more than R1 trillion over the current three-year period. Whichever measure one uses, the number is in the trillions, and the public balance sheet cannot carry it alone. On the savings side, the industry is substantial by any measure. The local collective investment schemes industry alone closed 2025 with R4.58 trillion under management, according to ASISA, and the broader savings and investment industry, spanning asset managers, life offices and retirement funds, holds considerably more.
Asset allocation: whose money is it?
The primary purpose of a retirement portfolio is to meet the long-dated, often inflation linked liabilities a fund owes its members. The job of the trustee and the manager is to build the portfolio that best matches those liabilities on a risk adjusted basis.
This reframes the infrastructure conversation in a way that matters. Infrastructure debt is not a charitable allocation. It is an asset class that earns its place in a well-built portfolio on its own merits. It offers contracted, often inflation-linked cash flows, long tenors that map neatly onto pension liabilities, and returns that are largely uncorrelated with listed equity and bonds.
Prescribed assets: why it should not be on the agenda
Prescription means the state directing funds to hold particular assets, whether government bonds, state-owned enterprise debt or specified projects, regardless of their risk adjusted merit.
Giving this legal force would not be a simple act of will. Regulation 28 is the instrument that governs how retirement funds may invest, and it currently contains no power to compel allocation to specified assets. It is our understanding that prescription would require the Minister of Finance to amend Regulation 28 itself, following a formal consultation process.
We are not supportive of it, and the reasons are practical rather than ideological.
The first is that the problem is misdiagnosed. The constraint on infrastructure investment in South Africa is not a lack of capital. It is a lack of bankable, well-structured and properly governed projects. The clearest evidence sits in the national accounts. South African gross fixed capital formation was running at just 14% of GDP in early 2026, well below the National Development Plan target of 30% and the 25% or more that economists typically regard as appropriate for a developing economy. The figure is also moving in the wrong direction: gross fixed capital formation fell by 2.7% over the first nine months of 2025 against the same period a year earlier, on Deloitte’s February 2026 reading of the Stats SA data, even as the economy returned to modest growth. For context, India invested close to 30% of GDP in fixed capital in the year ending March 2025 and China has remained around 40% in recent years. South Africa commits roughly half the share of its economy to building things that comparable emerging markets manage, and the gap is widening.
The regulatory headroom to direct far more towards infrastructure already exists. What is missing is not permission or money but a pipeline of projects that can be underwritten with confidence. Prescription treats a project-preparation failure as if it were a capital-supply failure, and so prescribes the wrong medicine. Compelling funds to allocate into a shortage of bankable assets does not create good projects. It simply forces savers to buy whatever is available, at whatever price, which is precisely the outcome a fiduciary exists to prevent.
The second reason is the cost to members. Directing savings towards assets at below-market returns, or towards entities that cannot price their own risk, produces sub-optimal outcomes for the savers whose money it is. That is difficult to reconcile with the fiduciary duty at the heart of the system.
The third reason is that better alternatives exist, built on de-risking rather than coercion. The Credit Guarantee Vehicle is the most prominent current example, though it is important to be accurate about its status. It is a plan in progress rather than a working facility. The World Bank board approved the supporting programme in early March 2026, and the vehicle is to be incorporated as an independent, privately governed non-life insurer, with an initial capital base in the region of R9 billion scaling towards a multi-year target of around R45 billion, subject to development partners confirming their participation. The Treasury is targeting operational readiness in the second half of 2026, with a first guarantee earmarked for independent transmission projects. If it delivers, the logic is sound. By providing market-based guarantees that reduce both the probability of default and the loss given default, it should widen the universe of investable projects and improve the risk-adjusted returns on those we finance, particularly in greenfield clean energy and transmission where construction and offtake risk are otherwise hard to price. The outcome will depend on the quality of the project pipeline, the efficiency of its processes and the commitment of partners to provide actual capital.
The common thread
Asset allocation, the regulatory limits and prescription are three versions of a single question about who decides where retirement savings go. The answer should be the trustees and managers who owe a duty to members.
The offshore allowance should stay, because diversification protects members.
Prescription should stay off the agenda, because it solves the wrong problem at the saver’s expense. And infrastructure deserves a far larger allocation than it currently receives, not because anyone is forced to provide it, but because, structured and de-risked correctly, it is simply a good investment.
