British economic growth accelerated to 0.7% in the third quarter of 2026, marking the strongest quarterly expansion since at least the start of 2025, and the Bank of England now faces a question it did not expect to be this awkward: is the economy strong enough to absorb another rate rise in six weeks, or is the headline number flattering a much narrower recovery underneath?
The Bull Case: Momentum Is Real, and the BoE Knows It
GDP accelerated from 0.5% in Q2 and is 1.9% higher than the same period in 2025, with September output rising 0.2% after no growth in August. That September figure matters: economists had forecast zero. Beating a flat expectation in a month widely assumed to be soft is not a rounding error.
The Bank of England last week upgraded its growth projection for 2026 to 1.6%, well ahead of the 1.2% forecast by the Office for Budget Responsibility following the October 28 budget. The central bank cited AI-related investment as one reason activity has been resilient, and it noted that some of the drivers of that resilience are not expected to persist. An institution that has just revised its own growth forecast upward and cited AI investment as a driver is not one that is hunting for reasons to pause.
Deputy Governor Clare Lombardelli has said she would not rule out a December hike. With annual inflation already at 3.5% in September and the Bank’s November forecast showing inflation peaking around 5% in the second quarter of 2027 under its central projection, a 0.7% growth quarter hands the MPC cover to move again.
The Bear Case: One Sector, Narrowing Fast
The problem sits inside the services surge. Computer programming and consultancy rose 4.4% and contributed 0.15 percentage points to overall GDP growth in the three months to September. That is roughly one-fifth of the entire quarterly gain from a single industry. The ONS says the industry has been the largest driver of GDP growth across 2026, and it has pointed to evidence that some of the biggest turnover jumps are tied to AI and cloud computing, even as it cautions it cannot measure that effect precisely.
Concentration risk is not theoretical here. Computer programming and consultancy output dropped 1.9% in September alone, raising questions about the sustainability of recent technology-led growth. A sector that powered Q3 and then contracted in its final month is not a foundation to build a tightening cycle on. Business investment grew just 0.4% in the quarter, which means the private sector beyond tech is not yet confident enough to commit capital at the rate the headline implies.
The same concentration question applies on the supply side of that AI billing cycle. Cloud infrastructure capex is expanding fast even as free cash flow turns sharply negative, which means the corporate spending underpinning UK tech revenues is itself under scrutiny.
There are signs of weakening consumer and business demand in the face of prolonged high energy prices and higher interest rates, and the October 28 budget has introduced its own fiscal drag. Tightening into a consumer squeeze funded by a single sector’s AI-related billing cycle is a reasonable place to hesitate.
Where the Evidence Leads
The bull and bear cases here are not really disagreeing about the data. They are disagreeing about how durable one sub-sector’s expansion is. The ONS itself acknowledges it cannot precisely quantify the AI contribution. That is an unusual foundation for a rate decision.
The stronger argument today is that the December hike is live but not certain. The sustained momentum strengthens the case for further tightening, but attention now shifts to October inflation data due on November 18, 2026 and the December meeting itself. That inflation reading, not today’s GDP, is the decisive input. If CPI surprises to the upside, market pricing can firm up quickly. If it softens, the BoE has genuine grounds to hold and assess whether the Q3 AI-linked surge carries into Q4 or fades as abruptly as September’s monthly data suggests it might.
The economy grew. The question is whether it grew broadly enough to justify tightening into it. Right now, the honest answer is: probably, but not certainly.
