$100 Million to $5 Billion in Unfunded Commitments: How Should Pension Funds Model Venture Capital Calls?
NVCA's 2026 Yearbook highlights why capital-call planning should not rely on quick exits. Venture exits improved in 2025 but remained far below the 2021 peak, while the unicorn backlog remained large. Capital calls can continue even when distributions lag.
NVCA reported $217 billion of U.S. venture-backed exit value in 2025, equal to 27% of the 2021 peak.
Build the Call Forecast Fund by Fund
A pension fund should model venture calls fund by fund, then combine them with every other private-market commitment and benefit payment. A $100 million programme and a $5 billion programme use the same logic but need different data, liquidity reserves, and governance. The model should show a base case, a faster-call case, and a no-distribution case. One expected curve is not enough.
| Unfunded commitments | Likely modelling need | Main risk |
|---|---|---|
| $100M | Fund-level schedule and liquid reserve | One large call can affect annual pacing |
| $1B | Vintage, manager, and strategy cohorts | Calls cluster across several funds |
| $5B | Integrated model with benefits, collateral, and total private markets | Small percentage errors become large dollar needs |
Suppose a pension has $1 billion of unfunded private-market commitments. A base forecast in which 30% is called over the next year requires $300 million of liquidity. If deployment accelerates and 40% is called, the requirement rises to $400 million. The extra $100 million may arrive during a period in which distributions are lower than expected and listed assets have also fallen.
The model should show which assets can fund that difference without disrupting benefit payments or forcing sales at an unattractive time. It should also identify which calls are contractual, which commitments can be paced more slowly, and how much liquidity remains after a downside case. A single average call rate cannot answer those questions.
Use the Actual Fund Calendar
New funds, mature funds, funds of funds, co-investments, and secondaries do not call capital the same way. The plan should use each manager's investment period, remaining commitment, expected follow-ons, and notice pattern. Historical behaviour from the same manager can improve the forecast, but a new strategy or larger fund may behave differently.
Pension liquidity exists to pay beneficiaries first. Venture calls should be tested alongside benefit outflows, public-market stress, derivative collateral, and other private-market calls. The hard case should not assume public assets can always be sold at normal prices or that distributions will offset calls.
The board should know what happens when coverage falls: which new commitments slow, whether co-investments pause, what liquid assets are used, and who can act between meetings. Existing commitments remain hard obligations. The response plan should protect them before discretionary new allocations.
Minimum Model Outputs
- Calls by quarter and year: Base, fast, and stress cases.
- Distributions by case: Including a severe delay.
- Liquid coverage: Assets available after benefits and other needs.
- Concentration: Largest manager and vintage call sources.
- Actions: Pacing changes tied to clear thresholds.
The purpose is not to predict every call. It is to make sure a large programme remains fundable when several assumptions go wrong at the same time.
If 30% of unfunded commitments are called in a stress year, $100 million of unfunded exposure creates $30 million of calls, $1 billion creates $300 million, and $5 billion creates $1.5 billion.
NVCA reported $217 billion of venture-backed exit value in 2025, but also described a large unicorn backlog, so distributions may not match capital-call timing.
Model Calls Against Liquid Assets
A pension plan with $5 billion of unfunded venture commitments and a 20% annual call assumption should reserve or source $1 billion of potential liquidity before counting distributions.
At a 30% call rate, $100 million, one billion, and $5 billion of unfunded commitments create $30 million, $300 million, and $1.5 billion of calls.
Capital Calls Under a 30% Stress Case
Unfunded commitments become a liquidity problem when call rates and weak distributions coincide.
View capital-call stress assumptions
| Unfunded commitments | Assumed annual call rate | Potential annual calls | Implication |
|---|---|---|---|
| $100M | 30% | $30M | Manageable if planned. |
| $1B | 30% | $300M | Requires portfolio cash planning. |
| $5B | 30% | $1.5B | Can affect total-fund liquidity. |
Pension funds can build durable venture portfolios, but manager count, commitment size, and timing need to match staff and consultant capacity. The allocation should be reviewed alongside buyout, growth, credit, real assets, and total-plan liquidity.
Capital calls do not wait for a convenient public-market environment. A pension can face calls from venture, buyout, real estate, and private credit at the same time that public assets have fallen and distributions have slowed. The forecast should therefore include a combined stress case rather than one model for each asset class. It should ask how much cash is required if calls arrive near the high end of expectations, distributions are delayed, and benefit payments continue as planned.
A large unfunded number is manageable when the plan has liquid coverage and time. A smaller number can be dangerous when it must be funded through forced sales. The useful measure is not unfunded commitments alone, but unfunded commitments relative to available liquid assets under stress.
Frequently Asked Questions
Are unfunded commitments the same as debt?
No, but they are liquidity obligations: The timing is uncertain, but a pension plan should assume managers can call capital under the fund documents.
Should distributions offset expected calls?
Only in a conservative scenario: Calls can arrive when distributions slow, so plans should model both separately.
Related Reading
unfunded commitment risk, commitment timing, and denominator effect.