Pre-MPC Briefing

100% AI Generated, No Human in the Loop

1. Executive summary

Thursday’s decision is a hold at 3.75%. Surveyed economist consensus is for a hold. August CPI, published 16 September, rose to 3.1% from 2.9% in July — in line with the Reuters poll, with core unchanged at 2.6% and services unchanged at 3.4%. The largest upward contribution came from transport, particularly motor fuels, so the post-July Brent surge is now visible in the consumer-price print rather than only in the oil price. That is a separate impulse from the July CPI rise, which was accounted for mainly by housing and household services via the Ofgem price cap. The three members who dissented for a 25 bp hike in July — Huw Pill, Catherine Mann and Megan Greene — receive reinforcement from both the oil move and Pill’s explicit 3 September remarks against “wait-and-see”. The remaining six members are expected to hold, producing a 6-3 split.

Gilts have sold off alongside the energy shock: the 10-year yield was 5.374% as of the 14 September close, 38 bp higher than at the 30 July 2026 meeting on the series used in Chart 32. SONIA forwards, as of that same gilt close, price only 4 bp of tightening for Thursday itself but 84 bp cumulatively by April 2027. QT is also on the agenda; the current £70 bn envelope (October 2025–September 2026) expires and the MPC must set the pace for the next annual period. A slower envelope would be consistent with the higher yield environment already priced.

2. Review of the previous MPC decision

At the 30 July 2026 meeting the MPC voted 6-3 to maintain Bank Rate at 3.75%. "Six members … preferred to maintain Bank Rate at this meeting". "Three members … preferred a 0.25 percentage point increase in Bank Rate at this meeting". "The Committee judges that the risks to the inflation outlook are tilted to the upside relative to the central projection in the July Monetary Policy Report".

The minutes noted that CPI inflation has fallen to 2.6% since the previous meeting. The July CPI print of 2.9% was published on 19 August after the meeting and was accounted for mainly by housing and household services rather than the post-meeting Brent move. The Committee continued to monitor the Middle East situation and how its impact propagates through the economy. Forward guidance remained data-dependent: the MPC stood ready to act if material second-round effects from the energy shock emerged, but judged that tighter financial conditions and ongoing labour-market loosening provided sufficient insurance for now. The July majority was explicitly a wait-and-see hold, not a dovish bloc.

3. Economic data evolution since the last meeting

The 16 September ONS package is the last inflation print before the vote. CPI rose to 3.1% in the year to August, from 2.9% in July, matching the Reuters consensus. Core is unchanged at 2.6%. Services are unchanged at 3.4%. Goods rose to 2.7% from 2.2%. Transport, particularly motor fuels, made the largest upward contribution to the change in the annual rate, so the post-30 July Brent move has begun to pass through. That is distinct from the July 2.9% print, which was accounted for mainly by housing and household services linked to the Ofgem price cap; furniture also added then, and transport pulled the other way.

Producer prices printed with CPI. Input PPI rose 6.1% in the year to August, up from a revised 5.8% in July (the June print in the latest vintage is 7.3%). Output PPI rose 3.7%, up from a revised 3.3% in July. Crude oil was the largest contribution to the annual input rate. Brent itself is shown as $109.51 as of the 9 September chart observation, not as 16 September’s cash price.

The 15 September labour-market release is now in the table: unemployment 4.9% in May-July, regular AWE 3.5%, employment rate 75.1%. PAYE payrolled employees were 30.2 million in August, down 26,000 on the month and 145,000 on the year. Vacancies were 702,000 in June-August, down 8,000 on the quarter. A flat LFS unemployment rate is therefore not the full slack picture: payrolls and vacancies have softened while pay growth has stabilised. Real wages remain positive (regular AWE 3.5% against headline CPI 3.1%). Growth indicators are soft: GDP rose 0.4% QoQ in Q2 after 0.6% in Q1, while official 3m/3m growth slowed from 0.6% in the three months to May to 0.4% in the three months to July. This tells the MPC that energy-led upside risks have intensified and have started to appear in goods and fuel prices, even as the domestic labour market and demand indicators remain consistent with a wait-and-see hold.

4. MPC member dashboard

Committee-member remarks. Dates are publication dates. Only Bailey 4 September and Pill 3 September were published after the 30 July decision; the earlier extracts are the last monetary-policy remarks on record for those members.

Andrew Bailey — Expected: hold

Bailey’s most recent monetary-policy-relevant remarks were delivered on 14 July (pre-dating the 30 July 2026 meeting) and described a “fairly soft economic setting activity-wise”. His 4 September speech on the institutional form of independent central banks contained no policy content. He now has 16 September’s August CPI in hand: headline 3.1% in line with consensus, core still 2.6%, services still 3.4%. Against that print plus the further $17.60 rise in Brent to the 9 September chart observation, the Governor is likely to keep the current data-dependent stance. Expect hold.

Clare Lombardelli — Expected: hold

Lombardelli has been silent since the 30 July 2026 meeting. She dissented hawkishly once (December 2025, voting hold against a cut majority) but has voted with the hold majority at every subsequent meeting, including the 6-3 July decision. The renewed oil shock and 16 September’s fuel-led CPI rise raise forward inflation risks that align with her earlier caution, yet core at 2.6%, unemployment at 4.9% in May-July and softening retail sales argue for continued monitoring rather than an immediate shift. Unlikely to move off hold.

Dave Ramsden — Expected: hold

Ramsden has been silent since the 30 July 2026 meeting. His voting record shows one dovish dissent (February 2026, voting cut against a hold majority) before aligning with the hold camp thereafter, including the July 6-3 outcome. Labour-market stability (unemployment 4.9% in May-July) and still-positive real wages support his past concern about demand weakness, but the oil move and the August PPI rebound add upside risk that he has previously acknowledged. He is unlikely to shift to a cut.

Sarah Breeden — Expected: hold

In her 30 June speech “Agents of change” Breeden addressed AI, cyber risks and operational resilience and did not discuss monetary policy. She has one prior dovish dissent (February 2026) but has voted hold at every meeting since, including the July decision. The 15 September labour-market package — unemployment 4.9% in May-July, regular pay 3.5%, PAYE down 26,000 in August — plus slowing 3m/3m GDP to 0.4% leaves her wait-and-see position intact. Unlikely to depart from hold on 16 September’s in-line CPI.

Huw Pill — Expected: hike

In “Homophones – Remarks by Huw Pill” on 3 September, Pill struck a firmly hawkish tone. Key lines:

“my preference has been for a modestly higher level of Bank Rate”
“my own response to this question has pointed to a need to raise Bank Rate to 4%”
“the underlying measure remains stuck meaningfully above 2%, with the risks around this projection being clearly to the upside”.

The $17.60 Brent increase since the 30 July 2026 meeting, the August fuel contribution to CPI, and his own consecutive dissents for a hike reinforce the case he has made since April. Expect another hawkish dissent.

Alan Taylor — Expected: hold

Taylor’s most recent monetary-policy speech on 23 June (pre-dates the 30 July 2026 meeting) emphasised rising slack, the need for an “extended hold” pending clarity, and the risk that “if inflation looks to undershoot, we may end up having to cut quickly”. That is a cut-if-slack-opens-up reaction function, not a generic wait-and-see. The data since then — unemployment still 4.9%, PAYE and vacancies softer, core still 2.6% — have not contradicted that view, even though the oil move raises forward risks. He is the dovish wing of a hold, not a hiker. Unlikely to shift without a clearer slack opening.

Catherine Mann — Expected: hike

Mann’s 2 July speech (pre-dates the 30 July 2026 meeting) explicitly shifted “towards a longer hold, and potentially a need to lean against that risk” of inflation persistence and cited “more upside risks to inflation compared to downside risks for activity”. She then dissented for a hike in July. The further Brent surge, sticky services at 3.4%, and 16 September’s fuel-led rise in headline CPI have moved in the direction her research highlighted. Expect her to repeat the hawkish dissent.

Megan Greene — Expected: hike

Greene’s 2 June speech (pre-dates the 30 July 2026 meeting) focused on repeated supply shocks and “a larger risk of second-round effects emerging through the price-setting than the wage-setting channel”. She has dissented for a hike at the last two meetings. The oil move since July, unchanged core at 2.6%, and the August PPI input rebound keep those risks live. Expect another hawkish dissent.

Swati Dhingra — Expected: hold

Dhingra’s 24 June speech (pre-dates the 30 July 2026 meeting) argued that monetary policy should “look through” temporary supply-driven relative-price adjustments to avoid unnecessary output costs and warned that tightening slows green investment. That look-through view is why she sits on the dovish wing of a hold rather than in the wait-and-see centre. She has a long dovish dissent history but voted with the July hold majority. Softening PAYE and vacancies support her supply-shock framing, even as the latest Brent move raises headline risks. Unlikely to move to a cut without further labour-market deterioration.

5. Outlook for inflation, growth, labour market

Headline CPI stands at 3.1% against the 2% target, 1.1 pp above it, while core is at 2.6%. The July rise was accounted for mainly by housing and household services via the Ofgem price cap. The post-30 July Brent move — $109.51 as of the 9 September chart observation — is a separate shock whose first consumer-price imprint is 16 September’s fuel contribution, with more pass-through still ahead. The July Monetary Policy Report’s adverse energy scenario, in which higher energy prices lead to more persistent inflation, is the right frame for that path; $109 Brent sits in the territory that scenario was written to test, even without restating the Report’s exact assumed energy trajectory here. Growth momentum has slowed: GDP rose 0.4% QoQ in Q2 after 0.6% in Q1, while the official 3m/3m rate has declined from 0.6% in the three months to May to 0.4% in the three months to July. The labour market remains loose on the LFS (unemployment 4.9% in May-July, employment 75.1%) and is softer on the timelier payroll and vacancy measures. Regular pay growth of 3.5% still produces a small positive real-wage gap against 3.1% CPI. The last minutes stated that “The Committee judges that the risks to the inflation outlook are tilted to the upside relative to the central projection in the July Monetary Policy Report”.

What would change our view
- The 16 September August CPI print did not hit the pre-release hike trigger (headline >3.2% or services >3.6%). The hold base case stands.
- September CPI services above 3.6%, or a further Brent spike sustained above $110 through October with no offsetting demand weakness → would increase the weight on second-round risks and lift the hike scenario.
- Vacancies and PAYE continuing to fall and unemployment rising above 5.1% in the next labour-market release → would tilt the balance toward an earlier cut.

6. Policy implications and expected decision

QT is co-equal on the agenda. The current £70 bn envelope runs to September 2026; Thursday’s decision will set the pace for the subsequent annual period. The gilt sell-off since 30 July (10-year yield up 38 bp to 5.374% as of the 14 September close on the Chart 32 series; the 4.996% starting point is that same series, not a generic par yield) already constitutes tighter financial conditions that the July majority cited as providing insurance against second-round effects. A slower envelope (market modal expectation around £50 bn, though this is not a Bank announcement) would be consistent with that reaction function and would limit further upward pressure on longer-dated yields. The operational decision on whether to continue active sales of long-dated gilts will be particularly watched given the 30-year yield at 5.893% as of the same close.

The July majority was explicitly wait-and-see at the centre, with Dhingra and Taylor as the dovish wing of that hold, not a single “mildly dovish” bloc. Reaction function remains data-dependent with emphasis on whether the latest energy impulse feeds into services prices or wage setting. We expect a hold with the same 6-3 split as July. The OIS curve as of 14 September prices only 4 bp of tightening for Thursday’s meeting and 84 bp cumulatively by April 2027. Our base-case hold is therefore slightly more dovish than the curve at the very front end but aligns once the three persistent hawkish dissents are taken into account. Guidance is likely to repeat that "the Committee will continue to monitor closely the situation in the Middle East and how its impact propagates through the economy" while retaining the upside-risk tilt.

7. What the market is pricing (no trade recommendations)

7a. The spot gilt curve

Source: Chart 29–Chart 33, yields as of the 14 September close. The curve has steepened to 135.6 bp between 10-year and 3-month.

7b. The OIS forward curve — meeting by meeting

Source: BoE OIS data as of 14 September. The curve first departs from the current 3.75% at the November meeting and prices steady tightening thereafter, reaching approximately 3.4 × 25 bp moves by April 2027.

7c. Where our call sits vs market pricing
We expect a 6-3 hold with the same three hawkish dissents as July. The OIS curve prices only 4 bp of tightening for Thursday itself, so the market is marginally more hawkish than our base case at the front end but consistent once the persistent minority is recognised. The cumulative 84 bp priced by April 2027 aligns with the upside-risk language the MPC has carried since the spring. Consensus risk is therefore skewed toward a slightly larger hawkish tilt in the minutes than the statement alone would suggest. The 25% hike scenario is our author view; the OIS-implied rough probability of a 25bp move this week is 17.6%.

8. Scenarios

• Hold — 70%: August CPI is now in at 3.1% / services 3.4%, below the pre-release hike trigger | Analogous to the 30 July 2026 meeting (6-3 hold with three persistent hawks) | Majority continues to view tighter financial conditions and labour-market slack as sufficient insurance while treating the fuel contribution as the start of pass-through, not yet a second-round break.
• 25bp hike — 25%: one extra member joins the July hawks | There is no clean UK analog for an actual hike in this cycle; July is the analog for the hold, not for a hike | Would require the Committee to lean against second-round risks now that fuel is in the CPI print.
• 25bp cut — 5%: Unemployment rises above 5.1% and 3m/3m GDP falls below 0.2% | Analogous to the February 2026 5-4 hold with four dovish dissents | Downside demand risks dominate and the oil move is judged transitory; probability kept low because the curve prices no easing and the MPC’s risk language remains upside-skewed.

9. Conclusion

Base case for Thursday’s decision is a 6-3 hold at 3.75% with Pill, Mann and Greene repeating their hawkish dissents. The 16 September August CPI print (3.1% in line, core 2.6%, services 3.4%) did not fire the hike trigger. The QT envelope for the next annual period will also be set; a reduction from the current £70 bn pace would be consistent with the higher yield environment. Three things to watch at the 12:00 announcement and in the minutes: (1) any strengthening of the upside-risk tilt language now that motor fuels have lifted the annual rate, (2) explicit reference to the interaction between higher gilt yields and the QT decision, and (3) whether the Committee signals that it will require more than one additional CPI print before reconsidering the current data-dependent stance.