In a country of 1.2 million people, two funds hold assets equal to more than half of GDP. The Public Service Pensions Fund (PSPF), which serves roughly 42,000 civil servants, manages assets that, as of March 2026, remained 43% invested domestically against a 30% statutory minimum.
The Eswatini National Provident Fund (ENPF), the country’s only mandatory retirement scheme and open to every formal-sector worker, has grown to E7.14 billion ($442.1 million) up 13.2% in the year to June 2025. Together, according to a 2025 World Bank technical assessment, Eswatini’s retirement fund sector alone holds assets worth 52.4% of GDP, a concentration of institutional capital unusual even by the standards of Southern Africa’s small, pension-heavy economies.
For most of that time, the two funds have operated on entirely different logics. PSPF is a defined benefit scheme, civil servants are promised a fixed pension calculated from salary and years served, with government picking up the risk if the fund falls short. ENPF is a defined contribution provident fund: members build up a balance and take it out as a lump sum, usually starting at age 50 even while still working. Now Eswatini is trying to turn the second model into something closer to the first, and the fight over how to do it is exposing real tension over who bears the cost.
The Conversion
The idea of converting ENPF into a national pension fund is not new draft legislation, the Eswatini National Pension Fund Bill, has circulated in Parliament in various forms for at least a decade. What has changed is momentum. Through 2025 and into 2026, ENPF ran a public campaign under the Siswati name Lidlelantfongeni combining stakeholder forums, a radio and TV program, and continental endorsement at the Africa Pension Funds and Retirement Summit in Casablanca, where CEO Futhi Tembe presented the conversion as a model for extending retirement security to informal workers.
As of May 2026, the Ministry of Labour and Social Security told Parliament the conversion “is ongoing,” with the Bill before the House of Legislature. The government has publicly backed the transformation. The same Parliamentary session surfaced a detail that gets less attention than the Bill itself: ENPF’s own Governing Board has been without normal functioning for some time, after Business Eswatini and other stakeholder representatives boycotted board meetings demanding the removal of its chairman. The Ministry’s Principal Secretary told MPs member funds remain safe, but confirmed the board is “not the way it should be” an odd backdrop for an institution asking the public to trust it with a bigger, more complex mandate.
The mechanics matter. Under the proposed model, ENPF would shift from paying a lump sum to paying a monthly income after retirement, the core request behind what unions have called, since the 1990s, the “27 demands” for stronger social security.
The new fund would also extend coverage to the roughly 200,000 Eswatini workers currently outside any retirement scheme: the self-employed, seasonal labourers, and domestic workers. Contributions would remain mandatory and capped, according to ENPF, at E400 ($24.77) per worker per month a cap specific to the new fund’s design, distinct from ENPF’s own statutory provident-fund ceiling, which rose separately to E430 ($26.63) in January 2026 under an unrelated regulation with the new pension fund coexisting alongside, not replacing, existing occupational schemes.
Where The Argument Actually Sits
The public framing from ENPF is about inclusion and dignity: replacing one-off payouts that members routinely deplete within a few years with an income that lasts. Nothing in the reporting gathered for this piece disputes that framing directly. What is disputed by PSPF’s leadership and by independent commentators is what the conversion does to the rest of the system, and specifically to PSPF.
“Don’t gamble with civil servants’ savings” — PSPF CEO Masotja Vilakati, addressing the Editors Forum in Mbabane on the risks of the ENPF conversion.
PSPF’s public position has hardened. Vilakati has warned that expanding ENPF into a pension fund would force PSPF to liquidate investments, retrench staff, and abandon job-creation projects effectively crippling its role in the domestic economy, and that diverting civil-servant contributions to the new scheme would leave the fund’s stability entirely undone. ENPF, for its part, has moved to reassure the public that the two funds will continue to operate independently, with Tembe telling editors that PSPF’s civil-service base and the new National Pension Fund’s private-sector, informal, and self-employed base would not compete for the same money.
Independent financial commentary has pushed further. Financial expert Sandile Mbhamali has warned that folding ENPF contributors into a system alongside a fund already carrying a deficit could deepen that shortfall and expose the government to unfunded liabilities that may need to be recognised on the state’s own balance sheet under international accounting rules. He has also flagged a governance gap: the absence of a published actuarial valuation or full public consultation behind the conversion.
Resolving The Deficit Number
Three figures for the size of PSPF’s funding shortfall have circulated in public debate over the past year, and at first glance they don’t agree. A 2025 World Bank technical assessment put PSPF’s deficit at E5.5 billion ($340.6 million) as of 31 March 2024, an 86.2% funding ratio. A leaked Cabinet document, reported by Swaziland News, cited PSPF’s own April 2025 actuarial valuation at an E5.1 billion ($315.8 million) deficit and 88% funding ratio. Both come from independent, methodologically transparent sources and sit close together, with the funding position improving slightly between the two dates.
Set against those is a widely repeated figure of “E10.2 billion” ($631.6 million), which Vilakati himself cited while addressing the Editors Forum in Mbabane in August 2025 nearly double the other two estimates. But the same report that carries that figure also quotes Vilakati giving PSPF’s underlying components directly: liabilities of E42.7 billion ($2.6 billion) against assets of E37.5 billion ($2.3 billion). Subtracting the two gives a deficit of E5.2 billion ($322.0 million), and a funding ratio of 87.8% not E10.2 billion. That reconstructed figure lines up almost exactly with both the World Bank’s E5.5 billion and the leaked valuation’s E5.1 billion.
The E10.2 billion figure appears to be a reporting or transcription error internal to that single account, inconsistent with the asset and liability numbers printed in the same piece. On the weight of three converging, cross-checkable data points against one outlier that doesn’t survive its own arithmetic, PSPF’s deficit is best treated, for now, as sitting in the E5–5.5 billion range, not E10 billion a distinction that matters given how directly the larger figure has been used to argue against the ENPF conversion.
Why The Deficit Number Isn’t Just Trivia
The World Bank’s technical note gives useful independent grounding for why PSPF’s funding position is fragile in the first place, separate from the ENPF fight. PSPF holds far more in domestic assets than regulation requires 43% locally invested against a 30% statutory minimum, a gap that has widened further through 2026 as the fund pursues a stated strategic target of 50% domestic exposure largely in government bonds and local loans that have underperformed offshore alternatives.
The World Bank’s modelling suggests PSPF could have been roughly E3 billion ($185.8 million) better funded based on the assessment’s modelling period, cutting its deficit in half, had it simply held the regulatory minimum in local assets rather than nearly 50% more. Part of PSPF’s shortfall, in other words, is not investment bad luck. It is a policy choice to over-allocate toward Eswatini itself a choice the fund’s own annual reporting frames explicitly as “balancing its fiduciary responsibility to members with its role in supporting national economic development,” language that puts the two goals side by side without resolving which one wins when they conflict.
That is the deeper version of the conversion debate. It is not just two funds negotiating membership overlap. It is a live argument about whether Eswatini’s largest pools of capital exist primarily to serve their members or to serve the state’s development priorities and PSPF’s own government-bond concentration, which the World Bank describes as creating a “circular” relationship between the fund’s own workers and the government’s own borrowing, suggests that argument has already been quietly settled in one direction for years.
What This Means For Capital Markets, Not Just Retirees
Eswatini’s retirement sector is the core of its long-term finance system bigger than insurance and investment-advisory assets combined, and the primary source of capital for the country’s small collective investment scheme and investment-adviser market, which the World Bank puts at E34.8 billion ($2.2 billion) in assets overseen, roughly 73% of it retirement-fund money.
The sector’s regulator, the Financial Services Regulatory Authority, has itself been assessed by the World Bank as only “partly implementing” four of ten core supervisory principles set by the International Organization of Pension Supervisors including adequate resourcing and transparency; as of mid-2023, a ten-person department was overseeing 175 licensed entities.
A 2023–24 collapse at Ecsponent Eswatini, an investment company serving retirement beneficiary funds, damaged confidence across the sector and triggered regulatory action in South Africa, a reminder that Eswatini’s capital market intermediaries are thin enough that a single failure has systemic reach. And the ENPF Bill’s decade-long limbo has itself become a drag on the private pension sector, according to the World Bank, as employers hold back from establishing new schemes while the shape of the mandatory system remains unsettled.
For anyone allocating capital in Eswatini or structuring retirement benefits for staff there, the practical read doesn’t wait for Parliament: underwrite PSPF’s shortfall at the E5–5.5 billion range that survives arithmetic, not the E10.2 billion figure still circulating in public debate, and treat a regulator supervising 175 entities with ten staff, plus one beneficiary fund already wiped out by a single collapse, as the standing level of oversight risk rather than a temporary gap. The Bill’s limbo is itself a cost, every year it stays unresolved is another year employers sit out setting up new schemes, so the safer planning assumption is that the current rules, not the reformed ones, are what apply for the foreseeable future.
