A weekly letter on the movements of state capital. Free, by email.
Every verified correction and every material revision to the record, in public. This page is the proof that the corrections process works.
| 2026-09-01 | HKMA Exchange Fund Investment PortfolioC8 corrected 2 to 0 — no SEC EDGAR registrant exists. Grade A to B. |
| 2026-09-01 | Future FundC8 corrected 2 to 0 — the Future Fund Management Agency has no EDGAR presence; the 13F-filing "Future Fund LLC" is an unrelated Illinois entity. Grade A to B. |
| 2026-09-01 | GIC Private LimitedC8 corrected 2 to 1 — GIC files 420 beneficial-ownership reports and zero 13F. Grade C to D. |
| 2026-09-01 | Massachusetts PRIM (PRIT Fund)C8 corrected 2 to 0 — no 13F filer on EDGAR. Grade A to B. |
| 2026-09-01 | Korean Teachers Credit UnionC8 corrected 2 to 0 — zero EDGAR filings of any form. Grade B to C. |
| 2026-09-01 | QICC8 corrected 2 to 0 — only QIC-sponsored fund vehicles appear on EDGAR (Form D); QIC itself files no 13F. Grade B to C. |
| 2026-09-01 | Kuwait Investment AuthorityC8 corrected 2 to 1 — KIA files SC 13D/13G only; the Kuwait Investment Office CIK holds a single 2016 Form 3. |
| 2026-09-01 | NZ Super FundC8 corrected 2 to 1 — the Guardians of NZ Super file SC 13G only. |
| 2026-09-01 | New York City Retirement SystemsC8 corrected 2 to 1 — the NYC Comptroller files only proxy-related PX14A6G/N on behalf of the funds. |
| 2026-09-01 | Public Institution for Social SecurityC8 corrected 2 to 1 — PIFSS files SC 13D and Forms 3/4 only. Its US arm Wafra files 13F but has been operationally independent since 2020 and is not credited. |
| 2026-09-01 | Qatar Investment AuthorityC8 corrected 0 to 1 — QIA does file SC 13D/13G. This correction moved the score UP. |
| 2026-09-01 | QICAssets figure corrected from a rounded "more than A$130bn" to the exact audited A$131.6bn at 30 June 2025. |
| 2026-09-01 | Emirates Investment AuthorityGrade corrected F (1/16) to D (5/16). We published a false adverse grade. EIA names its governing decree and all nine directors (C6 0 to 2) and publishes a min/max asset-allocation band table for eight asset classes on p.18 of its Inaugural Report (C7 0 to 2); its strategic portfolio is named (C4 0 to 1). It remains genuinely opaque on money - no financials, no AUM - but we scored it as though opacity about money meant opacity about everything. |
| 2026-09-01 | Libyan Investment AuthorityGrade corrected F (2/16) to D (7/16). We published a false adverse grade. LIA publishes a 59-page Annual Report 2024 with asset-class values and an equity book broken down by country, sector and currency (C4 0 to 1), an EY-examined 2019 financial statement (C2 0 to 1), an annual reporting cadence (C3 0 to 1), and its constituting statute Law 13/2010 in full with a named Board of Trustees (C6 1 to 2). |
| 2026-09-01 | Public Investment Fund (Saudi Arabia) updated to SAR 3.396tn ($905.6bn) gross AUM at 31 December 2025, from its 2025 Annual Report published 17 August 2026 — superseding the 2024 figure. PIF's gross AUM fell SAR 38bn during 2025, its first decline. The broader consolidated total-assets basis (SAR 4.54tn) is not comparable and is not used in the ranking. |
| 2026-09-01 | Correction — two false adverse grades withdrawn, and a scoring bias identified. The Emirates Investment Authority was published at F (1/16) and the Libyan Investment Authority at F (2/16). Both were wrong. Audited against their own documents, EIA is D (5/16) and LIA is D (7/16). EIA names its governing decree and all nine directors, and publishes a minimum and maximum allocation band for each of eight asset classes; LIA publishes a 59-page annual report with its equity book broken down by country, sector and currency, an EY-examined financial statement, and its constituting statute in full. The cause was not two isolated errors. We treated institutions that disclose nothing about money as though they disclosed nothing at all — and governance and strategy are precisely where otherwise-opaque funds do publish. Every remaining F and D grade is being re-audited on that suspicion, and those grades are the most adverse claims this publication makes. |
| 2026-09-01 | Correction — eleven filing scores revised after SEC EDGAR verification. Criterion C8 (regulatory filings) had been scored from assumption rather than checked against the register. Verified directly: HKMA and Australia's Future Fund have no EDGAR presence at all (C8 2→0, both A→B); GIC files 420 beneficial-ownership reports but zero 13F (2→1, C→D); KIA and NZ Super likewise file no 13F (2→1); QIA does file beneficial-ownership reports (0→1). The error was ours and is recorded here in full. A further 19 C8 scores are under verification and their grades may move. |
| 2026-09-01 | NSSF (China) updated to RMB 3,808.373bn total fund assets at 31 December 2025, from the Council's own 2025 annual report published 1 September 2026 (2025 return 13.22%), superseding the 2024 figure. Note that three separate pools are routinely conflated in reporting on China: the NSSF itself, the separately-managed entrusted basic pension fund (whose own report states it is independent of the NSSF), and the national social-insurance system balance. We rank the first only. |
| 2026-09-01 | Libyan Investment Authority updated to US$80bn total assets at 31 March 2026, from the LIA's own June 2026 statement, replacing a 2024 third-party citation. Mixed-date basis, disclosed on the entry: US$51.8bn of directly-managed assets at current market value, plus US$28.2bn of subsidiary holdings still carried at a 2019 valuation. |
| 2026-09-01 | QIC figure made exact. Replaced “more than A$130bn” with the audited A$131.6bn at 30 June 2025 from the annual report tabled in the Queensland Parliament. QIC's own website claims “more than A$135bn”; that figure is undated and unaudited and has deliberately not been adopted. |
| 2026-09-01 | Disclosure Grade rubric frozen as v1.0. The eight criteria, their definitions and the banding are fixed for edition 2026 and the full scoring function is now published on the methodology page. A grade now moves only because an institution's disclosure moved, never because the ruler changed. Future rubric changes ship as new numbered versions and are recorded here. |
| 2026-08-31 | First edition published: 100 institutions, $28.4T tracked, 87 primary-sourced figures, 13 attributed estimates, all 100 disclosure-graded. |
Represent one of these institutions — or know the numbers better than we do? Send corrections or newer figures with supporting documents. Verified updates are applied within one business day and credited in a public changelog — institutions that engage are how this record stays accurate.
Email a correction