Since the financial crisis, productivity growth has been stubbornly weak, acting as a drag on wages, living standards and economic growth. That's why the recent improvement in the data is worth paying attention to. The obvious suspect is AI. After all, if productivity is finally rising just as firms are embracing tools like Copilot and ChatGPT, it's natural to connect the dots. Perhaps, after years of working 9 to 5 and getting nowhere, Britain is finally getting more done. But the evidence points somewhere less glamorous: firms appear to be getting more output from workers they hired during the post-pandemic labour shortage rather than receiving a sudden boost from artificial intelligence.

Why UK productivity matters for economic growth and living standards

Productivity is essentially a measure of how much an economy produces for the amount of work that goes into producing it. The most common measure is output per hour worked: if workers can produce £100 worth of goods and services in an hour today, but £105 in an hour next year, productivity has increased by 5%. This can happen because workers have better technology or equipment, have acquired new skills, or simply because businesses have found more efficient ways of organising work.

Productivity matters because, over time, it is the main way an economy can raise living standards without requiring people to work longer hours. Indeed, Nobel Prize-winning economist Paul Krugman once quipped that “Productivity isn’t everything, but, in the long run, it is almost everything.” If workers can produce more in each hour, there is more income available to pay higher wages, more profits for businesses to invest, and more tax revenue to fund public services.

The UK's productivity problem is not a small one. Before the financial crisis, productivity was growing by about 1.8% per year. Since 2008, that has slowed to just 0.4% per year on average. Had productivity continued to grow at its pre-financial-crisis rate, the UK economy could now be around 30% more productive than it is today. Put another way, the amount of economic output available per person could be roughly £13,500 higher a year. That is not to say every person would literally have an extra £13,500 in their bank account, but it gives a sense of the enormous living-standard cost of Britain's productivity slowdown.

However, over the last two years productivity growth has averaged around 1% compared to about -0.5% in the previous two years. Admittedly, productivity is still far below its pre-pandemic path. There is one important caveat around the data. The traditional official labour market data has been plagued with issues, meaning we have to rely on alternative sources. But the most reliable data we have suggests a clear change in the direction of travel.

So, it’s good news that there is growing evidence that UK productivity has started trending up again over the last couple of years and, given the timing, it raises the obvious question: is this being driven by AI?

What is driving the recent UK productivity recovery

The first thing to note is that the productivity recovery is broad-based. Over the last two years, productivity has only fallen in four sectors: mining, construction, real estate and education. These are probably not that surprising given the drop in North Sea production and the dismal housing market over the last couple of years. So, it’s not that one sector is dragging up the rest.

The second thing is that there’s not much evidence that this is being driven by AI. Given ChatGPT was released at the end of 2022, adoption by businesses would have had to be impossibly quick for it to have started driving up productivity in 2025. Probably the best piece of evidence suggesting AI is driving the recent uptrend is that productivity in IT has accelerated at a much quicker rate than the rest of the economy. That would make sense as the IT sector tends to be at the cutting edge of AI implementation.

However, productivity growth in the IT sector has always been much faster than the rest of the economy and it was significantly faster in the two years before the pandemic than in 2024–2026. If it really was AI driving productivity growth in the sector, we would expect to see productivity accelerating above pre-pandemic trends, not slowing. What’s more, productivity growth in the finance, professional services and administration sectors, the other sectors where we would expect to see the impact of AI first, was slower than that of sectors like hospitality and transport, where AI should have a smaller impact.

What’s more, there is little evidence of AI replacing labour. Most firms surveyed by the Bank of England say AI has had “no material impact” on headcount over the past three years and the proportion of firms saying they are cutting jobs because of AI remains low.

In short, the evidence for an AI-driven productivity boom remains surprisingly thin. The productivity revival looks real. The AI revolution, at least in the economic data, still looks largely prospective.

Instead, the most likely explanation is a bit more mundane. Firms overhired coming out of the pandemic, when the labour market was tight. The number of people working rose by 2.6% between 2022 and 2024, while output grew by only 1.7%, meaning employment was growing faster than the economy. The extremely tight labour market coming out of the pandemic encouraged firms to “hoard” labour, even if they didn’t have enough work to properly occupy those workers.

Since 2024, though, output has grown by about 2.8%, while employment has grown by just 0.5% as hiring has dried up. In essence, firms have been making much better use of their current workforce before hiring more people. To put it in accountancy terms, utilisation has improved.

Is the UK productivity recovery sustainable?

The good news is that productivity appears to be improving again and, crucially, the recovery looks broad-based. Some of the recent gains are probably catch-up from the unusually weak productivity performance immediately after the pandemic, meaning the underlying trend is likely less impressive than the headline figures suggest. Even so, the UK’s productivity engine appears to be running a little faster than it was a few years ago.

Perhaps the most interesting implication is that if AI isn’t driving the recent improvement, then the economy may not yet have experienced the main event. The UK’s strengths in finance, professional services and technology mean it is unusually well placed to benefit from successful AI adoption. In other words, the recent productivity revival may have happened before AI has really begun to move the dial. For an economy that has spent much of the past fifteen years waiting for productivity growth to return, the most exciting possibility is that the biggest gains may still lie ahead.

It seems that the positivity from the sunny weather and a decent run in the world cup that we have seen in the consumer confidence data is also seeping into business sentiment. Almost all the business surveys that we follow have shrugged off the Iran war, another round of trade tariffs and a change in administration and have recovered to their pre-war levels.

That matters because just as more confident consumers are more likely to spend, more confident businesses as more likely to invest and hire, helping to grow the economy. There are also good signs that corporate borrowing is recovering as well, which is a good leading indicator of business investment picking up.

The UK has been notoriously poor at investment over the last few years, which is part of the reason for the underperformance of the economy. Business investment isn’t just good for growth, it is also one of the key ways to boost productivity and sustainable growth.

We expect the Final Composite PMI for August to confirm that the economy’s managed to sustain its solid momentum through the summer and that the pickup seen in July was more than just a temporary boost from England’s world cup performance.

Elsewhere, the Construction PMI, which doesn’t have a flash release, likely rose from 44.7 to 46.5 as new orders surged by 6.1 points in July and firms’ optimism about the outlook rose to the highest since the start of the Iran conflict, consistent with an improvement in the PMI. That said, the bigger picture is that the construction PMI has been below that crucial 50 no-change mark for over a year and a half now which will continue to pose a big headwind to Andy Burnham’s mission to build the most homes since the post-war period.

All told, we expect the Final PMIs to show that the economy managed to sustain its momentum in August despite persistent uncertainty and sticky inflation. Further ahead, we still think that growth will slow towards the end of the year as higher utility bills drag on household incomes, squeeze margins which means firms will have to contend with a softer demand backdrop as growth slows from its punchy 0.5% per quarter average in the first half of the year.

Related insights