Wycliffe Shamiah, CEO of Kenya’s Retirement Benefits Authority, is sitting with a problem that defines the next decade of African infrastructure finance. KEPFIC the Kenya Pension Funds Investment Consortium, built explicitly to channel domestic pension savings into infrastructure has over $1 billion theoretically available under its 10% regulatory cap. Only $113 million has moved. Only 88 of 1,075 eligible pension schemes have joined. The question Shamiah is answering right now is not whether Kenyan pension trustees want to invest in infrastructure. It is why the money stays in Treasury bills.
South Africa already answered that question. And the answer closes the debate.
The Regulation Was Never the Problem
In 2022, South Africa’s National Treasury amended Regulation 28 to allow pension funds to allocate up to 45% of assets to infrastructure. Three years later, actual allocation sits at roughly 2%. The gap between permitted and actual is not a few percentage points it is 43 points, representing hundreds of billions of rand that has legal clearance to move but does not.
This is the most important data point in African infrastructure finance. South Africa’s pension trustees manage $390 billion the largest pool of institutional capital on the continent, in a market with functioning bond markets, secondary trading, and sophisticated asset managers. They have the authority. They are not moving because the investable instruments do not exist in a form they can prudently hold.
GIZ confirmed the pattern across sub-Saharan Africa in March 2026: 1.1% of total private pension AUM (assets under management) allocated to alternatives, 0.5% to private capital. Pan-African infrastructure allocation averages 1 to 4%. This is not a regulatory problem. It is a product design problem.
Cross-reference: Africa Capital Markets: Tier Analysis and Market Readiness
What the Money Actually Needs
Three reform models that worked Chile, Malaysia, India share a common architecture that African markets have not yet built.
Chile’s AFP (mandatory private pension) system did not unlock infrastructure investment by raising regulatory caps. It did so through 40 years of institutional sequencing: mandatory savings scale first, independent regulatory credibility second, specific investable vehicles third. The 2024 reform explicitly channeled new employer contributions into “productive and infrastructure projects” but only because the prior architecture existed to receive them. Chile’s pension AUM reached approximately 80% of GDP precisely because the system was built before the asset classes were opened.
Malaysia’s EPF (Employees Provident Fund, $260 billion AUM) invests directly in toll roads, utilities, and airports. The precondition was not a regulatory amendment. It was Bank Negara’s independence and demonstrated currency stability. Trustees must know currency risk is bounded before they accept project risk on top of it.
India’s critical innovation was structural. The InvIT Infrastructure Investment Trust is a listed, liquid, income-generating vehicle that lets pension funds buy and sell infrastructure exposure rather than committing to 20-year project lock-ups. The illiquidity premium is not a behavioral bias to overcome. It is a legitimate fiduciary constraint. The InvIT resolves it by design.
The common sequencing across all three:
- Mandatory savings mandate AUM scale
- Independent regulator prudential framework trust
- Liquid secondary market infrastructure enable exit
- Specific infrastructure vehicle solve illiquidity
- Demonstration transactions with MDB (multilateral development bank) anchor build track record
African markets are stalling between steps three and four.
The Currency Layer Nobody Has Solved
Beneath the pipeline problem sits a structural gap that blended finance alone cannot close.
African infrastructure projects generate revenue in local currencies Kenyan shillings, Nigerian naira, Ghanaian cedis. When those projects are financed in USD, the currency mismatch becomes a sovereign contingent liability the moment the naira falls 60% (as it did in 2023) or the cedi requires IMF stabilization. A domestic pension fund holding local currency savings cannot rationally take on USD-denominated project risk. The currency mismatch defeats the purpose.
The TCX Fund a specialist hedging vehicle has hedged over $2.8 billion in local currency bonds across 28 currencies. The AfDB invested $25 million in TCX equity in September 2025 to expand Africa-specific hedging capacity. A March 2026 TCX-AfDB paper documents the case for local currency power purchase agreements: if energy offtake contracts are denominated in local currency and hedged through TCX, the entire project financing chain can be structured without USD exposure.
The binding gap: TCX cannot yet provide long-tenor hedging beyond 15 years. Infrastructure projects requiring 20 to 30-year amortization remain structurally mismatched. Until that gap closes, currency risk is a structural ceiling on how much domestic pension capital can realistically reach infrastructure regardless of regulatory caps or pipeline volume.
Cross-reference: TCX and Local Currency Infrastructure Finance: What African Structurers Need to Know
Who Can Move Now vs. Who Needs to Build First
Not all African markets face the same problem at the same severity. The OECD Africa Capital Markets Report 2025 and Oxford analysis identify three tiers.
Tier 1 Move now: South Africa, Namibia, and Botswana have functional bond markets, established secondary trading, and regulatory frameworks capable of hosting infrastructure issuance today. The constraint is purely pipeline and product design. An InvIT equivalent does not exist at scale in any of these markets. Fixing that is the highest-leverage intervention available.
Tier 2 Near-term potential: Kenya, Nigeria, and Mauritius have improving regulatory frameworks but lack the secondary bond market liquidity, rating agency coverage, and custody infrastructure needed to support institutional-grade infrastructure instruments. Nigeria carries an additional constraint: government bonds offering 15 to 22% nominal yields crowd out infrastructure bonds that cannot compete on yield without substantial credit enhancement.
Tier 3 Sequence first: Rwanda’s documented gap is instructive. The Capital Market Authority was established in 2011. The Rwanda Stock Exchange launched. RSSB, the national pension fund, is growing. But RSSB’s AUM is too small to anchor an infrastructure bond offering. The Milken Institute identified approximately $1 billion in AUM as the threshold for credible domestic anchor capacity. Rwanda is not there yet. The capital market cannot develop faster than the institutional savings base that anchors it.
Cross-reference: Institutional DNA Test: Applying the IDT Framework to African Capital Market Readiness
What MDB Leverage Math Actually Tells Us
Blended finance is a component of the solution. It is not the solution.
The AfDB’s Room2Run structure transferred mezzanine risk on a $1 billion loan portfolio to private investors, freeing AfDB balance sheet capacity for new lending without new donor capital. The EIB joint MDB report documents $3.0 billion in private finance mobilized from $1.6 billion in concessional commitments a 1.87:1 leverage ratio at aggregate level, rising to 6 to 8:1 for guarantee instruments.
What blended finance cannot do is substitute for structural architecture. A first-loss tranche does not create the secondary market. It does not solve the valuation methodology problem for pension fund actuaries. It does not provide the 25-year local currency hedge. These require institutional infrastructure built over years.
The demonstration transaction that Tier 1 African markets need to execute once, at scale: MDB first-loss plus local pension fund senior tranche plus TCX hedge plus listed liquid infrastructure vehicle. Kenya or South Africa are the candidates. The instruments exist in partial form. The architecture connecting them does not yet.
Scenarios
Demonstration Leads, Continent Follows (25%)
A single landmark transaction in Kenya or South Africa an InvIT-equivalent vehicle, TCX-hedged energy or transport deal, MDB first-loss tranche generates the proof of concept that pension fund trustees require. The template replicates to Tier 2 markets within five years. Infrastructure allocation in lead markets rises from 2% toward 8 to 12%.
Implications: Investors with early positioning in Tier 1 market infrastructure vehicles capture the premium before replication compresses it. Finance ministers who build the pipeline now, rather than advocating for regulatory cap increases, own the political credit for unlocking the capital.
Reform Without Architecture (50%)
Governments raise regulatory caps. Some mandate domestic infrastructure allocations. Without vehicle design, currency hedging infrastructure, and secondary market depth, allocation moves from 2% to 4% over five years. Capital remains concentrated in government bonds. The pipeline problem persists into the 2030s with periodic donor conference declarations about unlocking private capital.
Implications: MDB deal volumes remain constrained to demonstration transactions rather than scaling. Corporate strategists planning Africa infrastructure plays on domestic co-investment assumptions face timeline slippage of 5 to 10 years in most Tier 2 markets.
Currency Event Resets the Clock (25%)
A major currency dislocation in a Tier 2 market a Nigeria or Ghana equivalent combined with governance failures on one or more high-profile infrastructure projects, destroys nascent appetite for local infrastructure investment. MDB risk appetite retreats. Long-tenor local currency hedging development stalls. Timeline for meaningful domestic capital mobilization extends a decade.
Implications: The hard-currency financing model persists with its structural mismatch intact. Demonstration transactions shift to South Africa exclusively. TCX expansion pauses. Regional integration agendas ECOWAS, EAC capital market harmonization lose momentum.
Bottom Line
Africa holds more than $4 trillion in domestic capital. The binding constraint on infrastructure finance is not the regulatory cap South Africa’s 45% cap with 2% actual allocation settles that argument. The constraint is the absence of investable instruments that meet pension fund trustees’ legitimate fiduciary requirements: liquidity, transparent valuation, currency stability, and exit optionality.
The five-step sequencing model from Chile, Malaysia, and India is replicable. The currency hedging infrastructure exists in partial form. The MDB leverage toolkit is available. The demonstration transaction a listed, liquid, local-currency-denominated infrastructure vehicle anchored by an MDB first-loss needs to happen once in a Tier 1 African market.
The IMF and World Bank Spring Meetings this week are the right forum to demand that architecture, not another study on regulatory reform.
Why This Matters
“Africa’s pension funds are not allergic to infrastructure. They are allergic to infrastructure deals they cannot exit, cannot value, and cannot hedge. Fix the instrument, not the regulation.”
For African finance ministers and DRM (domestic resource mobilization) officials: Political capital spent advocating for higher regulatory caps is misallocated. South Africa already has a 45% cap. The agenda is pipeline development, vehicle design, and currency hedging architecture none of which requires new primary legislation, all of which requires deliberate institutional effort.
For MDB and DFI deal structurers: The leverage ratio of a guarantee instrument is meaningless if the underlying project cannot be held by a domestic pension fund. Designing the demonstration transaction in a Tier 1 market listed, liquid, local-currency, MDB first-loss is the highest-value single intervention available. The AfDB-TCX local currency PPA (power purchase agreement) framework published in March 2026 provides the hedging architecture. The vehicle design gap is what remains.
For corporate strategists entering Africa: Domestic pension capital is a realistic co-investor in African infrastructure, but only in markets with functioning secondary bond markets South Africa, Namibia, and near-term Kenya and only through instruments that structurally resolve the illiquidity premium. The capital is real. The timeline to instrument readiness is 3 to 7 years in Tier 2 markets. Size positions accordingly.
Key data sources: FSD Africa (January 2026), GIZ ICAMA Regulatory Assessment (March 2026), Pensions World SA (May 2025), Control Risks Kenya analysis (May 2025), TCX Fund LCY Bond data (2025), AfDB-TCX LCY PPA Paper (March 2026), OECD Africa Capital Markets Report (November 2025), CGD “Breaking the Bottlenecks” G20 Paper (March 2026).