Predictions

The Monetary Policy Committee has now held the policy rate at 11.5% for two consecutive meetings after its easing cycle paused. CPI inflation just surged to 11.15% y/y in August, more than tripling from 3.56% a year earlier, driven by a government fuel-cost pass-through policy, a 20.17% jump in transport costs, and a 13.89% rise in food prices (onions +46%, eggs +12%) tied to provincial wheat-supply failures. That surge is running well above the IMF programs 8.4% inflation forecast for the fiscal year, arguing against a cut. But SBP has historically preferred to observe further prints before reversing direction, real rates remain only modestly positive rather than deeply negative, and the rupee has stayed stable (around 277.5 to 278.65 per USD) with reserves near $17.1bn -- none of which forces an immediate hike. Hold is the more likely outcome; a hike is the main risk to this call if September data confirms the surge wasnt a one-off.

Headline CPI rose from a 15-month low of 2.6% (June) to 2.9% (July), exactly matching consensus. Underlying pressure is actually easing, not building: core CPI held flat at 2.6% y/y and services inflation cooled to 3.4% from 3.6%. The swing factor is fuel -- pump prices rose through early August then eased again by mid-month, a choppier picture than Julys clean base-effect-driven jump. The Bank of Englands own guidance is that inflation stays just under 3% for most of 2026, with the acceleration toward a ~3.25% Q4 peak not expected yet. Call: roughly flat with July, not a repeat of June-to-Julys full 0.3pp jump.

The Fed held at 3.50% to 3.75% in July on a divided 9-3 vote. Since then the data has turned: July payrolls fell -23,000 with May and June revised sharply lower, marking two straight months of soft labor data, while unemployment held at 4.1%. July CPI cooled to 3.4% y/y from 3.5% in June, the second consecutive month of deceleration, with core at 2.5%. Market pricing already reflects this shift: roughly 85% probability of a 25bp cut priced in ahead of the meeting. Two more data points land before the decision -- the August jobs report (4 Sept) and August CPI (11 Sept) -- and could still move the calculus, but the weight of evidence points to a cut rather than a hold.

Q1 2026 grew +0.6% q/q. The Q2 monthly path was uneven: April fell -0.1% (revised), May was flat at 0% (revised down from +0.1%), June rebounded +0.3% on services strength. ONSs own rolling comparison -- growth in the three months to June 2026 vs the three months to March 2026 -- already came in at +0.4%, down from +0.6% for the three months to May, and that comparison window is methodologically almost identical to the quarterly release itself. Consistent with the IMFs 1.0% full-year 2026 growth forecast, which implies a slower pace than Q1 across the rest of the year.

August inflation jumped to 11.15% y/y from 9.2% in July, driven substantially by a one-off government fuel-cost and tax pass-through plus a wheat-supply-driven food price shock (onions alone up 46% m/m). Because the fuel policy move was a discrete action rather than an ongoing trend, and agricultural price spikes like onions tend to be volatile rather than persistent, some moderation is plausible. But underlying food-supply pressure and transport costs look more structural, and rural inflation (12.2%) is running well ahead of urban (10.4%), suggesting broad-based rather than narrow pressure. Net call: roughly holding near Augusts level rather than a sharp reversal either direction. Note: PBS does not publish an advance release calendar the way ONS does; the early-October date is inferred from its consistent month-start release pattern, not officially confirmed.

Cross-checked against 5 indicators: flash composite PMI up sharply to 52.1 (bullish); retail sales -0.5% m/m, read as payback from May/June promo pull-forward rather than new weakness (bearish but discounted); energy price cap +13% from 1 July (genuine headwind); GfK consumer confidence +6pts, biggest jump since Nov 2023, rising despite the energy shock (bullish); unemployment flat at 4.9%, claimant count fell 2nd straight month (neutral). No analyst consensus was publicly available at time of writing; professional consensus historically misses by ~0.2pp.