Bank of England: AI added 0.5% to productivity in three years

The Bank of England just put a number on AI productivity, and it is small. In its November 2026 Monetary Policy Report, published November 5, the Bank estimates that AI has lifted UK productivity by about 0.5% over the past three years. The Register reports that this works out to 0.1 to 0.2 percentage points of extra productivity growth a year since early 2024. The interesting part is not the size. It is the mechanism: firms are getting the gain by hiring fewer people, not by firing them.

What actually happened

Per The Register's reading of the report, the Bank sees "only evidence of a modest AI-driven rise in UK productivity growth." Businesses in its Decision Maker Panel survey report "virtually no overall impact on employment so far." But AI makes some employers less likely to hire new staff or to replace people who leave. Output keeps growing while hiring slows. That is where the productivity gain comes from.

Firms expect more. They forecast AI will add around 0.9 percentage points to annual productivity growth over the next three years, split roughly equally between faster output and slower employment growth. Those are the firms' own estimates, not measured results.

The Bank also estimates that AI-related activity accounts for just under a fifth of UK GDP growth since the start of 2026. Most of that comes from building and selling AI, not from other companies using it. The same report raised Bank Rate by 0.25 points to 4%, on an 8–1 vote.

Why it matters for your business

This is the macro picture catching up with what we see on the ground. AI rarely deletes a job on a small team. It removes the next hire. The support inbox grows 30% and you don't add a second rep. The bookkeeper leaves and the replacement is part-time.

That gain is real, but it does not show up on any report you already run. Headcount stays flat, so nobody gets credit. If you want to know whether your AI tools pay for themselves, track the things that move:

• Hours per unit of work. Tickets per rep-hour, orders processed per hour, invoices closed per week.
• The hire you didn't make. Write down the role you would have opened without automation, and what it would have cost.
• Where freed hours go. Time saved and spent on sales is a gain. Time saved and lost to reviewing AI output is not.

We have said this before: don't wait for a central bank to tell you whether AI works. Now one has, and the answer is "slowly, through hiring."

Sources: Bank of England, Monetary Policy Report November 2026, The Register.