UNIVERSAL ASSET OWNERS | RESPONSE-LED ANALYSIS
Germany's nuclear-waste fund must coordinate €2.778 billion of uncalled commitments with annual public payments as it raises unlisted assets from 14.3% to 29% - without new contributions.
By UAO Editorial Team | 6 August 2026
KENFO has one pool of money and two calendars it does not fully control. Private-market managers call committed capital on their schedules. Germany’s nuclear-waste program presents reimbursement bills on another.
The fund must meet both without fresh contributions. At the same time, it plans to increase unlisted assets from 14.3% of the portfolio at the end of 2025 to 29% by 2028.
The goal is not a simple portfolio rebalance. KENFO’s investment overview sets the 29% target, while its 2025 annual report records €2.778 billion of open commitments that may be called before the build-out ends. The test is whether a finite fund can accept those calls without being forced to sell liquid assets at the wrong moment.
In a written response to Universal Asset Owners, Finn Maximilian Strickert, a project manager at KENFO, described the same strategy in rounded terms: “Our strategic asset allocation targets approximately 30% in illiquid assets, alongside roughly 32% in equities and 38% in fixed income.” The public target is 29%, and the audited 2025 figures—not the rounded email mix—remain the basis for the allocation calculations below.
No refill, and no pause in the bills
KENFO is a public-law foundation created in 2017 to invest funds transferred by Germany’s nuclear-power operators and reimburse the federal government for specified waste-management costs. Its starting capital was €24.1 billion. Beyond that one-time transfer, KENFO says it receives no further contributions from the operators or the federal government.
The mandate therefore differs from an open pension plan collecting regular contributions or a sovereign fund receiving recurring commodity revenue. Investment returns replenish KENFO’s capital; scheduled reimbursements reduce it. The fund says its horizon extends to the end of the century, while its strategic asset-allocation framework links the assets to roughly 75 years of expected obligations.
The audited 2025 annual report puts the investment portfolio’s year-end market value at €25.577 billion. Of that, €3.663 billion was in unlisted assets: infrastructure, private equity, private debt, real estate and a small “other” category. That is 14.32% by calculation, consistent with the reported 14.3%.
Strickert said KENFO deliberately delayed significant real-estate investment during the 2020 pandemic disruption and the 2022 inflation shock, then began making meaningful real-estate investments in 2023. He described private equity as about 39% of capital invested in the illiquid portfolio. Real estate and infrastructure remain below their target volumes and are therefore KENFO’s current deployment focus, he said; that pacing does not signal a material change in portfolio weights, and private equity remains an important part of the private-markets portfolio. The 39% figure is not the same denominator as the annual report’s 34.4% private-equity share by market value, so the two percentages should not be compared as if they measured the same thing.
Why 25.2% is not an allocation
At Dec. 31, 2025, KENFO also had €2.778 billion of open commitments to unlisted investments. Adding that amount to the €3.663 billion already invested produces €6.441 billion, equal to about 25.2% of the €25.577 billion portfolio.
That ratio is a useful measure of the build-out pipeline, but it is not a 25.2% asset allocation. An unfunded commitment is an obligation to supply cash when a manager calls it. Until then, it has no current market value in the unlisted portfolio; the timing and amount of calls can differ from the headline commitment, while distributions can offset gross cash demand. The denominator will also change with returns and reimbursements.
The distinction matters because a commitment can make progress toward 2028 look nearly complete while increasing future demand on liquid assets. KENFO’s public website says amounts awaiting investment are held in replacement investments in equities and bonds. That keeps capital working, but it also means a capital call can require cash to be raised from a market portfolio whose value may be moving at the same time.
Strickert explained the bridge between commitment and call this way: “Capital earmarked for illiquid investments is held in liquid assets with a comparable risk profile until it is called.” He added: “KENFO's structure differs from most sovereign funds in that it cannot rely on ongoing inflows; our asset allocation reflects this constraint.”
The relevant governance question is therefore not only how much has been committed. It is how calls are forecast, how quickly liquid holdings can be converted, which assets are reserved for the next government payment and what happens when those demands coincide with weak markets. The same choreography confronts mature pension, endowment and sovereign vehicles building private-market exposure against fixed spending or liability calendars.
The 2025 cash choreography
The annual report shows the mechanism in operation. KENFO made €870 million of new private-market commitments in 2025. Managers called about €719 million, while the unlisted portfolio distributed about €266 million back to KENFO. The €453 million difference is a simple measure of calls less capital distributions; it is not an investment gain or the change in market value.
Calls were below KENFO’s expectations, partly because weaker merger-and-acquisition markets slowed activity in underlying funds. The year-end unlisted share of 14.3% also fell short of KENFO’s expected 15.6% to 16.6% range by 1.3 to 2.3 percentage points. A weaker US dollar against the euro reduced the reported value of dollar-denominated assets as well.
Slower calls ease immediate cash demand but defer the build. If activity later accelerates across several funds, the same commitments can create a more concentrated need for liquidity. Distributions may offset that need, but their timing is also uncertain. That is why commitments belong in a liquidity plan even though they do not belong in invested market value.
The other cash calendar was larger. KENFO reimbursed €823 million of disposal costs in 2025. Its annual report says €635 million of distributions from its master fund, together with the cash flow from redeeming units with a €325 million market value, supplied the necessary liquidity. It also says all reimbursement amounts were paid without delay.
Those transactions show how KENFO met the 2025 payment, but they do not disclose a permanent minimum buffer, the distribution of future calls by month or a simultaneous-call stress result. Those remain important limits on what an outsider can conclude about liquidity adequacy.
Returns do not settle the liability question
KENFO reported a 6.2% total return on its foundation assets in 2025, above the 3.97% target listed in the annual report’s year-end risk section. Its current performance page separately shows a 6.4% return on invested capital, a narrower denominator. In a July 15 release, KENFO reported a further 7.9% gain for the first half of 2026 and said invested assets were then about €28 billion. Strong returns expand the capital base, but they do not establish that a multi-decade funding plan is sufficient; total foundation assets lost 12.2% in 2022.
The target also moves. KENFO defines it as the average annual return needed to cover projected disposal costs through 2099; higher cost estimates or market losses raise the required rate. Its risk framework calls this Deckungsrisiko, or coverage risk, and says assets, payment profiles and cost estimates are tested through recurring asset-liability studies and annual long-term planning.
An independent countercase comes from Mahdi Awawda and Alexander Wimmers. In a 2026 Energy Policy study, they modeled delays in repository siting and the associated extension of storage costs. Across the paper’s two delayed-pathway bundles, the full results table reports minimum average returns of 5.54% to 6.63%; the de-facto-delay bundle alone ranges from 5.91% to 6.35%, compared with the 3.7% target used in the model.
Those figures are scenarios, not a forecast of KENFO’s actual shortfall. The authors say their cost base comes from a 2015 study using an outdated repository design, their base inflation assumption is 1.6%, and their work does not evaluate KENFO’s investment strategy. Their accepted manuscript also shows the results are highly sensitive to inflation and project timing. The study’s value here is to expose the second uncertainty: private-market pacing can be managed inside the portfolio, but the liability estimate can move independently of investment execution.
The four numbers that matter
For KENFO’s board and public overseers, the 29% target should therefore be read with four separate measures: unlisted market value, uncalled commitments, the call-and-distribution schedule, and the disposal-payment schedule. Collapsing them into one “private-markets exposure” number hides the cash problem the strategy must solve.
The observable tests through 2028 are concrete. The invested unlisted share should move toward the target without commitments being relabeled as assets. Calls and distributions should remain reconcilable with changes in the liquid portfolio. KENFO should continue to make annual reimbursements without delay. And revisions to the target return should show whether higher disposal-cost estimates, market performance or both are changing the burden.
Reaching 29% will not prove the liability is fully funded, just as remaining below it does not prove the strategy has failed. The more demanding result is a portfolio that can accept calls when they arrive, make public payments when they fall due and preserve enough liquid risk capacity to avoid turning a long horizon into a forced short-term decision.