UK productivity growth in 2026 is showing signs of something economists have spent years waiting for: a meaningful improvement in output per hour.

New analysis from the Resolution Foundation suggests output per hour has grown at an average annual rate of around 1.1% over the past two years. That is a significant change from the 0.7% annual decline recorded over the preceding two-year period.

The conclusion fits with experimental work from the Office for National Statistics, which is increasingly using administrative payroll data to address weaknesses in the Labour Force Survey. Productivity determines how quickly companies can raise wages and profits without generating the same pressure on prices.

The measurement problem has been obscuring the recovery

Britain's productivity debate has been complicated by problems with labour-market statistics. Post-pandemic response rates to the Labour Force Survey deteriorated, making estimates of hours worked less reliable.

That matters because productivity is calculated partly by comparing output with labour input. If the hours-worked measure is wrong, the productivity calculation can also be misleading.

The ONS now recommends paying close attention to an approach that makes greater use of PAYE Real Time Information data. Its latest flash estimate suggests output per hour was 0.7% higher in the second quarter of 2026 than a year earlier, while output per worker increased 1.4%.

The gains appear broader than an AI boom

It would be tempting to credit the entire improvement to artificial intelligence. British companies are investing heavily in software, automation and AI tools, particularly across professional services, financial services and technology.

But the Resolution Foundation argues that the improvement looks broad-based. The gains do not appear to be explained simply by a collapse in low-productivity employment or a narrow technology boom.

A productivity recovery built only around a small group of AI-intensive companies would have much less impact on national living standards than gains spread across the wider economy — one of the tests set for the UK's industrial strategy.

Why businesses should care

Productivity is ultimately about producing more value from the same amount of labour and capital. For an individual company, that can mean automating administrative work, improving software, redesigning processes or investing in equipment.

When productivity rises, businesses have more room to increase salaries without simply passing the entire cost into prices.

Wage pressure remains significant, while the Bank of England is alert to the risk that rising labour costs keep services inflation elevated. Stronger productivity is one way of breaking that link, and it sits alongside the wider UK growth picture.

AI's real test is now diffusion

The AI investment boom will still matter. The question is whether productivity gains move beyond technology teams and into normal operating functions.

A law firm using AI to draft first-pass documents, a retailer improving demand forecasting or a manufacturer using software to reduce downtime can all raise output without hiring proportionately more people. That diffusion is already visible in how UK firms are adapting to data and AI regulation while deploying tools in production.

The strongest productivity cycles usually occur when technology stops being novel and becomes routine.

It is too early to declare victory

Britain's post-2008 productivity record remains weak. Two years of better data do not reverse more than a decade of underperformance, and statistical uncertainty means the current improvement should be treated carefully.

But administrative data, private-sector estimates and recent economic resilience are beginning to point the same way.

If AI is contributing, its most valuable effect may be happening quietly inside ordinary businesses rather than in the companies selling the technology.