South Africa has one of the deepest institutional savings pools in the emerging world. Retirement fund assets now approach R7-trillion, up from the R5.84-trillion last measured across the industry by the Financial Sector Conduct Authority in 2023. This gives South Africa an assets-to-GDP ratio that exceeds that of many Organisation for Economic Co-operation and Development (OECD) economies, making it Africa’s only globally significant pension market. Yet gross fixed capital formation has fallen to about 15% of GDP, well below the 30%–35% investment rates that underpinned the rapid growth of economies such as South Korea, China and Vietnam. One of the world’s richest pools of patient capital co-exists with one of its weakest investment economies. The usual diagnosis is that pension funds are too conservative, regulation too restrictive or trustees too risk averse. But South Africa does not suffer from a shortage of capital. It suffers from an institutional intermediation gap, the absence of organisations capable of systematically transforming long-term savings into productive investment. That distinction matters because it changes the diagnosis and the remedy. The retirement system is behaving exactly as it was designed to. Its primary purpose is to preserve workers’ savings through prudent governance, diversification and fiduciary discipline. That means favouring liquid listed equities, government bonds and offshore assets over illiquid infrastructure projects, industrial investments and early-stage enterprises whose risks are harder to price and justify. The outcome is predictable. Retirement savings finance ownership of existing assets far more effectively than the creation of new productive ones. Pension funds own a substantial share of the JSE, but they finance relatively little new fixed capital. The transmission mechanism between savings and investment breaks down in the following places: Fiduciary incentives. Infrastructure and private markets are permitted under regulation 28, but trustees must justify every allocation on a risk-adjusted basis. Too many projects simply fail that test. Country risk. Years of policy uncertainty, deteriorating infrastructure and underperforming state-owned enterprises have raised the risk premium on domestic investment. Trustees are responding rationally, not irrationally, when they limit exposure. South Africa lacks a consistent pipeline of bankable projects. This diagnosis is increasingly echoed across sectors: Business Day recently reported that from energy to logistics investors describe capital as available, but well-prepared, investment-ready projects are in short supply. The prescribed assets debate assumes pension funds are refusing to finance investment-ready opportunities. More often, too few projects reach institutional standards of preparation, governance and predictable cash flow. The fourth constraint is the most important because it helps explain the persistence of the first three. South Africa has built institutions to manage savings, but not the institutional machinery required to convert development opportunities into investable assets. Countries that successfully mobilise pension capital rarely expect pension funds to originate, structure and manage infrastructure investments themselves. Instead, they rely on specialist intermediaries that prepare projects, structure transactions, allocate risk and package investments into products suitable for long-term institutional investors. Canada’s public pension managers built sophisticated internal investment capabilities that enabled them to become global infrastructure investors. Australia’s superannuation system relies heavily on specialist investment managers such as IFM Investors. Denmark’s Copenhagen Infrastructure Partners packages infrastructure into fiduciary-grade investment products, while Britain’s National Wealth Fund provides guarantees and catalytic capital that make projects institutionally investable. In none of these systems does a pension fund buy a toll road. It buys exposure to an institution whose purpose is to transform infrastructure into an investable asset class. South Africa has elements of this ecosystem, but they do not yet form a coherent whole. The Public Investment Corporation manages public-sector assets rather than originating investment products for the wider retirement industry. Meanwhile, the Development Bank of Southern Africa (DBSA) and the Industrial Development Corporation (IDC) finance strategically important projects, but together their balance sheets total about R265bn, less than 4% of South Africa’s almost R7-trillion retirement savings pool. Together, they remain too small to bridge South Africa’s investment gap. What South Africa lacks is an institutional platform dedicated to preparing projects, pooling risk and creating investment products specifically designed for long-term retirement capital. One possible model would be an independently governed South African infrastructure investment platform. Rather than asking pension funds to evaluate individual greenfield projects, such an institution would prepare projects to institutional standards, combine them into diversified portfolios and use development finance to absorb risks that retirement funds cannot reasonably carry. Its mandate would differ from the DBSA’s or IDC’s in kind, not just scale. Instead of lending from its own balance sheet, its purpose would be to manufacture investable assets for pension funds and insurers. The result would be a new asset class rather than a collection of individual projects. Trustees would not be asked to finance a single transmission line, port or water scheme. They would invest in diversified, professionally managed portfolios designed to meet fiduciary requirements while expanding productive investment. Viewed through this lens, the prescribed-assets debate largely misses the point. Compelling pension funds to invest in projects that are not institutionally investable does not repair the broken transmission mechanism. It merely transfers development risk onto retirement beneficiaries while weakening the governance principles that have made South Africa’s retirement system one of the strongest in the developing world. The real challenge is institutional rather than regulatory. Instead of asking why pension funds are not investing more aggressively, policymakers should ask a different question: who is responsible for systematically transforming development opportunities into fiduciary-grade investment products? The policy implications are straightforward. Build and capitalise the intermediary layer. Institutionalise project preparation. Reduce sovereign and policy risk through credible long-term commitments. Use development finance strategically to absorb risks that long-term retirement capital cannot. Above all, stop asking trustees to be braver. Start by building institutions worthy of their confidence. South Africa does not lack savings. It lacks the institutions capable of converting savings into productive investment. Until that missing layer is built, the country will remain rich in long-term capital but poor in long-term investment. The question is no longer where the capital is. The question is why South Africa has not yet built the institutions capable of putting that capital to productive work. • Skosana is MD and head of public affairs for South Africa at FTI Consulting. He writes in his personal capacity.Business Day