Three separately-graded sub-claims for the August 5 RBI decision — repo rate, stance, and the vote split, each a probability partition. The edge isn't the (likely) hold; it's that the risk has quietly flipped from dovish to hawkish, so the residual surprise is a hawkish dissent, not a dovish one.
Stance: Neutral retained 0.95. Vote: 6-0 unanimous 0.72. We say plainly we're consensus on the hold — the edge is the stance language and the dissent direction: the dovish-dissent faction has dissolved, so the residual surprise is hawkish, not dovish.
We lock three independent sub-claims, each a probability partition summing to 100%, each read directly from the RBI Resolution.
| Sub-claim | Locked probabilities |
|---|---|
| Repo rate | Hold 5.25% 0.91 · Cut 0.03 · Hike 0.06 |
| Stance | Neutral retained 0.95 · Withdrawal of Accommodation 0.04 · Accommodative 0.01 |
| Vote split | 6-0 unanimous 0.72 · 1 dissent 0.23 · 2+ dissents 0.05 |
The cutting cycle has paused after 125bp of 2025 cuts (Feb 25 + Apr 25 + Jun 50 + Dec 25), with the repo at 5.25%. CPI is rising (May 3.93%, fifth straight) and the June meeting raised the FY27 CPI path to 5.1% (Q3 peak 5.9%) while cutting GDP to 6.6%. Oil has fallen ~33% from its 2026 peak (~$119.50 → ~$80.59), defusing the West Asia energy risk the RBI cited. Manufacturing PMI is solid at 55.0 (May). The single biggest swing input before the meeting is the June CPI print on ~July 12–13.
Sources: RBI June-2026 statement (Republic World, Deccan Herald); Business Standard (125bp cuts, Dec 2025); CNBC / TradingEconomics (oil); Business Upturn (PMI 55.0).
Through 2025 every dissent leaned dovish — at October 2025, two external members wanted an accommodative stance. That faction has dissolved: by April and June 2026 the votes were fully unanimous, the cut was delivered, and the macro flipped (rising CPI, oil shock in the pipeline). So the August tail risk is no longer a surprise dovish cut — it is a hawkish dissent or withdrawal-of-accommodation language if the July CPI prints hot. A dovish-tilted book is fighting the new data.
The P-02 position-desk method, applied to the six MPC members (three RBI, three external) — each member's likely vote modelled from their public record and the data they've cited, then aggregated into the three partitions. The stance partition uses the RBI's official vocabulary only: Neutral / Withdrawal of Accommodation / Accommodative.
A hot July CPI (>5%) raises P(hawkish dissent) and could nudge the stance to Withdrawal of Accommodation. A soft print (<3.5%) revives the cut case and a dovish dissent. A cut would be the full miss on the rate partition. We pre-commit these so resolution is read against the literal Resolution, never spun.
RAOSCAFF locks P-04 on 2026-06-20, before the Aug 3–5 meeting. Each sub-claim resolves to a discrete published value in the RBI Resolution — the repo rate (a number), the stance (a verbatim official term), and the vote tally (an integer) — read directly from one authoritative document on a fixed date.
Official RBI stance vocabulary only — Neutral / Withdrawal of Accommodation / Accommodative — so there is no 'what we meant' ambiguity at resolution.