Update: IMF Quota Consents Stuck at 72.8% — Still −12.2 pp Short of 85%; US Not Filed

What changed since the research vintage

In our institutions & governance research ledger we mapped the stock of formal authority: the United States alone holds about 16.5% of IMF votes — enough to block 85% special majorities — China’s PPP GDP share (~18.7%) still dwarfs its ~6.1% Fund vote weight, and the 16th General Review of Quotas raised total quotas by 50% while leaving relative shares frozen. That post answered the design question — how do power structures allocate authority? This update answers the effectiveness question the newest official prints force: what moved in the consent and shareholding vintages, and did any lever actually reallocate votes?

Two documents rewrite the timeline without rewriting the hierarchy. The IMF Finance Department’s Policy Paper 2025/040 (published November 2025) discloses that as of 29 October 2025, only 132 members representing 72.78% of quotas had consented to their 16th GRQ increases — 12.22 percentage points short of the 85% effectiveness threshold — and that the Executive Board extended the consent window to 15 May 2026. The World Bank’s DC2026-0003 report to Governors (10 April 2026) closes the 2025 Shareholding Review: 45 of 189 IBRD members (holding 47.5% of shareholding) are under-represented under the Dynamic Formula, yet there is not sufficient support for a Selective Capital Increase (75% needed) or a Basic Votes increase (85% Articles threshold). The dashboard above is built as a vintage delta: consent shortfalls, the path to the May 2026 deadline, IBRD misalignment, frozen vote−GDP gaps, and which reform levers moved versus stuck.

The headline table: research stock → newest print

Read the table as a freeze with a clock, not a quiet period. Money was supposed to arrive (equiproportional quota increase). Authority was not supposed to move. The update says even the money has not cleared the consent gate — and the Bank’s five-year realignment ritual ended without issuing shares.

The binding shortfall is consent, not formula math

Toggle the dashboard’s Consent track control. The quota bar sits at 72.78% against an 85% line; the NAB bar at 83.90% against 90%. Those are not forecasting confidence intervals. They are hard institutional rules in Resolution 79-1 and the companion NAB decision: no member’s increase becomes effective until the Board determines that the thresholds are met.

The arithmetic of the remaining gap is stark. Annex I of PP 2025/040 states that consents representing a further 12.22% of total quotas are still required. The United States is absent from the consented list. In our research print the US alone sits near 16.5% of votes and roughly 17% of quotas — larger than the entire remaining shortfall. China (6.40% quota share), Japan (6.47%), Germany (5.59%), France and the United Kingdom (4.23% each), India (2.75%), and Brazil (2.32%) have consented. The coalition that can finish the review on paper already includes most of the top tier — except the shareholder whose vote share is the structural veto on 85% decisions in the first place.

That is the vintage punchline for the theme question. Authority is allocated not only by charter formulas but by who must say yes before formulas become cash. An equiproportional increase that preserves US blocking power still cannot take effect without US (and other pending) consent. The research post showed why relative shares did not move; the update shows why the increase itself remains contingent.

World Bank review: misalignment measured, realignment refused

Switch to the IBRD misalignment donut. Under the Dynamic Formula (80% GDP blend, 20% IDA contributions, with compression), 45 countries accounting for 47.5% of shareholding are under-represented; 144 countries with 52.5% of shareholding are over-represented. Low-income countries hold about 2% of IBRD voting power. The Review’s own language is blunt: after simulations of Selective Capital Increases, Basic Votes hikes, and targeted unallocated-share options, there is not sufficient support for an SCI (Governors’ resolution needing 75%) or for raising Basic Votes (Articles amendment needing 85%).

Compare that to the 2020 review, which also produced no shareholding adjustment. The 2025 vintage is not a surprise failure; it is a confirmed pattern. The formula can print misalignment every five years. The majorities required to fix it sit inside the same weighted-voting structure that the misalignment describes. Pair this Bank ledger with our NATO defense-spending vs GDP and SIPRI world military expenditure posts when you need the security-budget side of institutional power — here the binding constraint is shareholding arithmetic, not tanks.

What did move is the Voice track: an expanding Voice Secondment Program, Research Analyst posts for the largest multi-country chairs, a LIC Board working group, formalized President-selection process tweaks, stronger client access at Spring/Annual Meetings, and Global South think-tank partnerships. Those are real procedural gains. They are also explicitly budget-neutral and non-share. The dashboard’s voice-package split makes the hierarchy visible: soft voice advances; hard vote realignment defers.

Gaps did not budge — consent status is the only new color

Filter Region on the scatter. Every major’s IMF vote − PPP GDP gap is identical to the research print: China still −12.6 pp, India −5.3 pp, Indonesia −1.6 pp, Japan and Germany still over-weighted. The Δ column is a row of zeros because relative shares never moved — and because the 16th GRQ was designed not to move them. What the scatter adds is consent coloring: dark markers for non-consenters, teal/region hues for those already on the October list. The geometry of under-representation is unchanged; the path-dependence of effectiveness is newly measurable.

That matters for readers who treat “the 16th Review happened in 2023” as settled. Governors approved the Resolution. Effectiveness is a second gate. Until 85% of quotas consent and 90% of NAB credit arrangements consent to the rollback, the Fund’s permanent resource mix stays on the old rails even though the political bargain was struck. Deadline extensions — including the latest jump to 15 May 2026 — are how the institution buys time without admitting the bargain is incomplete.

Who is exposed under the new vintage

Exposed: emerging-market coalitions that treated the 16th GRQ’s +50% headline as money in the bank; reformers who assumed the Bank’s 2025 review would deliver even a modest SCI after the 2020 no-op; program countries whose access narratives embed larger Fund quota resources that have not yet become effective; and anyone modeling “quota share = current voting power dynamics” without a consent overlay.

Relative winners under current rules: shareholders who already sit above special-majority blocking thresholds and can treat consent timing as leverage; chairs that extract voice concessions (secondments, analysts, LIC working groups) while voting weights stay put; and bilateral or plurilateral safety-net arrangements that do not wait on Fund effectiveness conditions.

What would change the story: US (and remaining) quota consents that clear the 12.22 pp gap before the May 2026 deadline; NAB participants covering the last 6.10 pp to 90%; a surprise SCI coalition at the Bank clearing 75%; or a 17th GRQ that finally abandons equiproportional freezes. None of those print in PP 2025/040 or DC2026-0003.

Caveats and methodology

• Consent ≠ ratification politics alone. Some members cannot consent for domestic or recognition reasons (PP 2025/040 notes Afghanistan, Myanmar, and Venezuela as unable to consent at this stage). The US absence is still the largest single disclosed gap relative to the shortfall size.
• Quota share ≠ vote share. Consent thresholds use quota denominators as of 7 November 2023; voting power includes basic votes. We flag US quota % as estimated where the annex lists consented members only.
• NAB participant list ≠ IMF membership. Rollback consents are among NAB creditors; Japan’s ~18.4% NAB share dwarfs most peers. US NAB non-consent is coded from absence on the disclosed consented table.
• IBRD misalignment follows the Bank’s Dynamic Formula with 2022–2024 GDP and IDA through IDA22not identical to our research post’s PPP-GDP gap lens.
• “Moved” voice reforms are Board-recommended packages, not yet fully implemented outcomes; treat them as process wins, not completed institutional redesign.
• Vote−GDP gaps are carried from the research vintage because relative shares are unchanged; they are not a new WEO re-estimation.
• This update is a vintage delta. For the full authority-layer organogram (UNSC, chairs, veto series), use the research post.

The shareable takeaway

Versus our institutions research vintage, the newest official prints say authority still has not been reallocated — and the agreed quota enlargement is itself unfinished. IMF quota consents sit at 72.78%, −12.22 pp short of 85%; NAB rollback consents at 83.90%, −6.10 pp short of 90%; the United States is not on the October 2025 consent list; China’s −12.6 pp IMF vote−GDP gap is unchanged; and the World Bank’s 2025 review counts 45 under-represented members holding 47.5% of shares while refusing both SCI and Basic Votes. Soft voice measures advance. Hard voting weights stay frozen. In this system, power is allocated twice — once in the charter, and again in who must consent before the charter’s resource deal becomes real.

Related reading: Institutions & governance research and China all-budget fiscal revenue.