Same fund, two owners, different numbers — where LP disclosures diverge
Across the 31 plans in our private market performance tool (6,653 disclosures, 995 funds held by more than one plan), CalPERS and CalSTRS are the deepest pair to compare — and where they hold the same fund, the reported numbers don't always agree. We matched 85 funds held by both plans and compared their since-inception net IRRs. Most agree closely — the median gap across 52 mature funds is 0.73 points — but the outliers reach 7.19 points on the same underlying fund. That spread is the point: two investors in the same fund can both be reporting honestly and still print different numbers.
Part of Private market performance
The aggregate numbers behind Private market performance — for the live view, open the interactive tool.
Open Private market performance →Every figure carries its source and as-of date — quote freely, with attribution.
Pensions: all tracked plans · Australian super in numbers · Private market performance · the PE statistics · what plans disclose · where they diverge
Why would the same fund show two different IRRs?
Three honest reasons, all visible in the plans' own footnotes. The plans report to different dates (CalPERS as of December 31, 2025, CalSTRS as of June 30, 2025) — and because a private fund's value is only restated periodically, one plan can still be carrying a mark the other has already moved past. They may hold different vehicles of the fund, with different fee terms and therefore different net returns. And each computes its IRR from its own cash-flow dates, on its own method — CalSTRS notes in its report that there is no industry-standardised way to do it.
One thing that is usually not the reason: the valuation itself. A pension plan generally carries the fund manager's own reported net asset value rather than marking the holding independently, adjusting only where something has clearly moved that the manager's books have not yet caught up with. So most of a gap is timing and terms — not two investors disagreeing about what the same asset is worth.
The widest gaps — mature funds only (vintage 2021 or earlier)
Ranked by the size of the gap, not by vintage. Funds newer than 2021 are deliberately left out: early-life IRRs are not meaningful (the J-curve), so a gap between two young numbers means nothing yet.
| Fund | Vintage | CalPERS net IRR | CalSTRS net IRR | Gap (pts) | CalPERS multiple | CalSTRS multiple (derived) |
|---|---|---|---|---|---|---|
| Carlyle Asia Partners V, L.P. | 2018 | 15.2% | 8.01% | +7.19 | 1.3x | 1.21x |
| Carlyle Europe Partners V, S.C.Sp. | 2019 | 3.6% | -3.49% | +7.09 | 1.1x | 0.91x |
| Valor Equity Partners IV L.P. | 2017 | 25.9% | 20.19% | +5.71 | 4.8x | 3.17x |
| Carlyle Partners VII, L.P. | 2018 | 12.9% | 8.32% | +4.58 | 1.5x | 1.39x |
| WCAS XIII, L.P. | 2019 | 15.5% | 18.73% | -3.23 | 1.7x | 1.61x |
| Advent International GPE IX Limited Partnership | 2019 | 12.3% | 14.75% | -2.45 | 1.6x | 1.66x |
| Permira Europe III | 2004 | 26.6% | 24.15% | +2.45 | 1.7x | 1.66x |
| Carlyle Partners VI, L.P. | 2013 | 15.8% | 13.36% | +2.44 | 2x | 1.87x |
| Palladium Equity Partners V, L.P. | 2018 | 12.7% | 14.86% | -2.16 | 1.6x | 1.68x |
| Blackstone Capital Partners VII, L.P. | 2016 | 15.1% | 13.1% | +2.0 | 1.8x | 1.73x |
| Francisco Partners Agility II, L.P. | 2020 | 26.7% | 28.67% | -1.97 | 2.2x | 2.13x |
| Siris Partners IV, L.P. | 2019 | 6.1% | 8% | -1.9 | 1.3x | 1.31x |
| Hellman & Friedman Capital Partners IX, L.P. | 2020 | 12.4% | 14% | -1.6 | 1.7x | 1.72x |
| Centerbridge Capital Partners III, L.P. | 2015 | 12.6% | 14.12% | -1.52 | 1.5x | 1.58x |
| Ares Corporate Opportunities Fund V, L.P. | 2017 | 4.6% | 5.97% | -1.37 | 1.2x | 1.3x |
Top 15 gaps of 52 mature matched funds · young vintages excluded (early IRRs are not meaningful — both plans say so in their own footnotes).
This page shows the widest gaps — behind it sits the full matched dataset: every fund held by both plans, both sets of reported figures, and the deltas. Email us if you're interested in the full dataset.
How the matching works
Deliberately conservative: fund names are matched exactly after normalising style (punctuation, "&" vs "and", entity suffixes), plus a short hand-checked alias list. Parallel vehicles and sleeves of the same fund (a "Fund VI-A", an "L.P.2") are not treated as the same disclosure — they genuinely carry different fee terms. A name match whose vintage years disagree is rejected. Funds we could not match with certainty stay unmatched — 85 matches is what survives that bar, and CalSTRS' multiple is derived from its published cash flows (the report has no multiple column).
The underlying data: the private market performance tool (pick a plan, or compare plans side by side) · all tracked plans.