August Concentration: IMF Top-1 16.5% / Top-3 28.7% Frozen — Clock at 87 Days

Our Q3 concentration print answered the mid-year distribution question: how thick is formal authority at the tip when the consent gate moves — IMF Top-1 16.5%, Top-3 28.7%, Top-5 40%, with quota consents climbing +3.88 pp to 76.66%. Our August mid-window update then answered the vintage question: what moved after that print — consent flat, Diriyah principles on the books, 87 days left to the November gate. This August concentration lens asks the sharper follow-up desks trade next: how concentrated is the system when the tip is frozen, the consent meters stop, and the only mover is the calendar?

The interactive dashboard above is built as an August 202608 concentration lens. Toggle Scoreboard, Vote ladder, Clock & consent, and Block & veto. The punchline is deliberately three-sided. On voting power, the tip is still stuck: US 16.5%, Top-3 28.7%, Top-5 40%, IBRD nearly identical (15.8% / 28.5% / ~40%) — Δ vs Q3 = 0 pp. On consent, the gate is flat: quota still 76.66%, residual to 85% still −8.34 pp, NAB still 83.9%. On clock, optionality compresses: 54.5% of the Board-extended window is already spent; 87 days remain to 15 Nov 2026. Concentration did not dilute. Consent did not advance. The calendar did the only work.

The August concentration scoreboard

Read the table as a family of market shares plus a clock, not one slogan. Vote Top-1 / Top-3 tell you who owns continuous weight. Consent percentages tell you whether that weight can activate a reform. The clock tells you how much calendar remains to close the residual without a superseding consent print. Blocking arithmetic tells you who can stop the hardest votes even when most members say yes. Diriyah tells you that a reform language exists — not that Top-1 moved. Analysts who quote only “Top-5 = 40%” miss the solo block; analysts who quote only “consent = 76.66%” miss that the residual gap is still sized like a Top-1 shareholder; analysts who ignore the 87-day clock miss that flat consent is itself a concentration risk.

Frozen tip: Top-1 / Top-3 still 16.5% / 28.7%

Toggle Vote ladder. The Fund and Bank have not published a relative-share rewrite between the Q3 concentration print and this August mid-window. The Lorenz panel among the top ten still rises past one-sixth at rank 1 and past half by rank 9–10, while an equal-split diagonal across ten chairs climbs linearly. Ranked bars still put the United States alone near 16.5% IMF / 15.8% IBRD, with Japan and China clustered near 6% each — and China still sitting roughly −12.6 pp under its PPP GDP claim on the scatter.

That freeze is the August contribution relative to a naive “update = new shares” instinct. Our 2026 concentration companion scored the tip; Q3 re-scored it against a moving consent gate; August re-scores it against a stopped consent gate and a moving clock. What August rewrites is not the tip — it is the optionality around the tip: how little calendar remains to make an equiproportional enlargement effective while the Top-1 shareholder still withholds consent.

Open Scoreboard → Cross-perimeter HHI. Analytical HHI proxies still rank UNSC veto equality highest (~2,000 on five equal rights), then seat-bloc geometry, then the IMF/IBRD top-ten ladders (~520 / ~490). Continuous multilateral voting looks “mild” next to commodity chokepoints — until the 85% rule turns 16.5% into a structural veto on the hardest decisions. The August overlay does not change those HHI levels; it changes how expensive delay becomes when they sit next to a flat consent residual.

Flat consent: the gate stopped moving

Toggle Clock & consent. In Q3, the story was ascent: quota consents climbed from 72.78% (Oct 2025) to 76.66% (Apr 2026), narrowing the gap to 85% by almost 4 pp. By 20 Aug 2026, that ascent has stalled. No superseding IMF consent ledger has printed. Quota remains 76.66%. NAB remains 83.9%. Members consented on quotas still 149; pending still 42. The residual to effectiveness is still −8.34 pp on quotas and −6.1 pp on NAB — gaps that, in vote-weight terms, still look like a Top-1–scale holdout problem rather than a long tail of micro-states.

The path chart now carries four vintages: Oct 2025 → Apr 2026 → Aug mid → Nov gate. Switch the track to Days to gate and the August point becomes the visual hinge: days remaining drop from 191 at the May Board extension to 87 at the mid-window clock, while the consent level line is flat. That geometry is the August punchline in one pane — concentration without consent, under a compressing calendar.

China has consented; the United States has not. That is not a moral claim. It is arithmetic. The sole shareholder above the 15% special-majority block line is also the pivotal weight needed to close most of the remaining quota gap. Our institutions research ledger already mapped the vote-vs-GDP gaps that make Asia-Pacific under-weight on Fund chairs; August concentration simply notes that those gaps cannot clear a consent gate the Top-1 chair has not entered.

Clock compression is the August concentration story

The Board extended the consent period to 15 Nov 2026. From the May extension start to this mid-window print, 104 of 191 calendar days have elapsed — 54.5% of the window. 87 days remain. Flat consent inside a burning window is not “no news.” It is a regime: the tip stays thick, the gate stays short, and the option set for a late rush shrinks every week.

Scoreboard row Consent-window clock is therefore a first-class concentration perimeter in this vintage, not a decorative countdown. When vote Top-1 is frozen and consent Δ is zero, the only continuous series desks can update without a new Fund PP is the day count. Treat 87 as a risk meter on the same page as 16.5% and 76.66% — not as a separate ops calendar.

What would change the story mid-window: a superseding consent PP that clears a material slice of the 8.34 pp quota gap; a US filing; a surprise SCI coalition at the Bank that moves IBRD hierarchy talk; or a 17th GRQ package that turns Diriyah Principle 5 into a formula that actually moves shares. None of those print in the August sources. Until they do, the concentration answer stays: tip frozen, gate flat, clock burning.

Diriyah does not dilute the tip

The August update already logged the Diriyah Guiding Principles (PP 2026/013): eight principles on quota and governance reform, with Principle 8 still open on voice and representation. This concentration lens asks the narrower question: did Diriyah move Top-1 / Top-3?

No. Principle 1 (adequacy of quota resources) is about Fund size, not redistribution. Principle 5 (realignment toward economic weight) is conditional — it would need a 17th GRQ formula that actually relocates shares. Principle 8 remains open. The dashboard Diriyah table flags tip impact as none / conditional / open for that reason. Reform language without a share rewrite leaves the Lorenz curve where Q3 left it. Desks that treat Diriyah as a de-concentration event are reading a principles paper as a voting-power table.

IBRD stock as of 30 Jun 2026 confirms the twin hierarchy: US still ~15.8%, Top-3 still ~28.5%, SCI still without US-scale support in the August vintage. Bank-side concentration and Fund-side concentration remain cousins — both tip-heavy, both consent-gated, both unmoved by mid-window calendar compression.

Block & veto: practice still concentrates inside formal equality

Toggle Block & veto. Charter concentration already says five states own 100% of vetoes. Practice concentrates further: recent annual stacks remain dominated by Russia, with the United States and China as secondary users, and the United Kingdom and France near zero in the 2018–2024 window. Formal P5 equality (“each has a veto”) coexists with highly unequal use. Expanding elected seats without touching the veto still leaves Top-5 veto share at 100%.

On the Fund side, blocking arithmetic is unchanged from Q3: US alone clears the 15% line; Japan+China at 12.2% remain short; an EU-5 cluster can clear it as a bloc; an EM-5 cluster sits just shy. Consent flags on those clusters still matter: the solo blocker has not consented; several under-weight Asia-Pacific chairs have. That is why concentration and consent stay coupled even when consent levels stop moving — the identity of the residual holdout is the concentration story.

Caveats and method notes

• This is an August concentration lens. For the frozen tip without the mid-window clock overlay, use the Q3 concentration print. For consent clock mechanics and Diriyah narrative without Top-k re-scoring, use the August update. For vote-vs-GDP stock detail, use the research ledger.
• Relative IMF/IBRD vote shares are carried pending 16th GRQ effectiveness. Do not invent a mid-window share rewrite.
• Consent percentages are carried from PP 2026/017 into the August clock; “flat” means no superseding ledger by 20 Aug 2026, not a new Fund disclosure of zero change.
• HHI and Lorenz among the top ten are analytical derivatives, not Fund-published indices. They are for cross-perimeter comparison, not regulatory citation.
• UNSC veto-use counts are from public Security Council documentation and are illustrative of practice concentration, not a claim about legitimacy.
• PPP GDP anchors are WEO-style gaps rounded to 0.1 pp; they are gap meters, not exact quota formulas.
• Diriyah tip-impact labels (none / conditional / open) are editorial mappings from principles language to concentration outcomesnot Fund staff classifications.

What would change the story

Three coincident signals would force a rewrite of this August concentration print: (1) a superseding consent PP that moves quota share materially toward 85% or NAB toward 90%; (2) a US 16th GRQ consent (or a disclosed coalition that closes the residual without it); (3) a 17th GRQ / SCI package that actually relocates Top-1 / Top-3 weights rather than restating principles. Until one of those prints, the live August answer is clear: IMF Top-1 stays 16.5%, Top-3 stays 28.7%, consent stays 76.66% with an 8.34 pp gap, and the clock is down to 87 days.

Related reading: Q3 concentration print, August institutions update, 2026 concentration print, and institutions research ledger.