Productivity is the amount of output the economy produces for each unit of work put in, and it is the main long-run driver of living standards. This article explains how the Office for National Statistics (ONS) measures it, why its latest estimate depends on which data source is used, and how to read the figures.
What is productivity, and how does the ONS define it?
According to the ONS, labour productivity is calculated by dividing output by labour input. Output is measured as gross value added (GVA), which estimates the volume of goods and services produced. Labour input can be counted as workers, jobs or hours worked, and the ONS describes hours worked as the preferred measure.
That gives two headline versions:
- Output per hour worked compares output with the total hours people spend working.
- Output per worker compares output with the number of people in work, regardless of how many hours each of them works.
The difference matters. If people shift to shorter working weeks, output per worker can look weak while output per hour looks healthy. For that reason, output per hour is usually the cleaner guide to how efficiently time is being used.
What did the latest ONS release show?
The ONS published its productivity flash estimate and overview on 18 August 2026, covering April to June 2026 (the second quarter) and January to March 2026. The next labour productivity release is scheduled for 17 November 2026.
According to the ONS, output per hour was 0.7% higher in the year to the second quarter of 2026 when measured mainly using PAYE Real Time Information (RTI) data. Output per worker was 1.4% higher on the same basis. Using Labour Force Survey (LFS) sources, output per hour was 0.2% lower than a year earlier.
| Measure, year to Q2 2026 | RTI-based | LFS-based |
|---|---|---|
| Output per hour | +0.7% | -0.2% |
| Output per worker | +1.4% | +0.4% |
The ONS describes the overall picture as weak. Productivity has recovered from the disruption of the pandemic, but growth remains subdued compared with the stronger trends seen before the 2008 financial crisis. The ONS recommends the RTI-based measure as the best guide to recent changes.
Why do two sources give different answers?
Both sources use the same estimate of output. In the ONS release, GVA was 1.2% higher than a year earlier in the second quarter of 2026. The gap comes from the labour side of the calculation.
The ONS explains that RTI does not collect actual hours worked. Instead, hours for both sources are calculated by multiplying the LFS measure of average hours by each source’s count of workers. So if one source shows more workers than the other, it also shows more hours, and output per hour looks lower.
The ONS also points to changes in the Labour Force Survey since January 2024. It says recent movements reflect both the economy and better survey quality, including more responses from people in employment, so recent LFS employment growth is likely to be temporarily higher and LFS productivity growth temporarily lower. It is also investigating possible employment-related non-response bias, with initial LFS findings due in autumn 2026. For background on how the labour market data are built, see what employment figures can and cannot tell us.
How does the arithmetic work? A worked example
Productivity growth is not simply output growth minus hours growth, but for small changes the two are close. The illustrative arithmetic below uses the ONS second-quarter figures for the LFS-based measure, where GVA rose 1.2% and, according to the ONS, hours worked rose 1.4% over the year.
- Take an index of output at 100 a year earlier. It becomes 101.2.
- Take an index of hours at 100 a year earlier. It becomes 101.4.
- Output per hour moves from 100 ÷ 100 = 1.000 to 101.2 ÷ 101.4 = 0.998.
- That is a fall of about 0.2%, matching the published LFS-based figure.
The lesson is that output can grow and productivity can still fall. If hours rise faster than the value produced, each hour yields slightly less. Equally, an economy can post weak growth while productivity improves, if fewer hours are used to produce the same result.
Why does productivity matter for living standards?
Over long periods, an economy can raise average incomes in only two ways: put more hours of work in, or get more out of each hour. Population and employment rates cannot keep rising indefinitely, so sustained improvements in pay and consumption ultimately depend on productivity.
Higher productivity gives employers the capacity to pay more without raising prices, and gives governments more taxable activity for the same effort. It does not guarantee those gains are shared evenly or passed on in wages quickly, but without it the room for real pay growth is limited.
The ONS also notes a reallocation effect: the gain to the whole economy when activity shifts from lower-productivity to higher-productivity industries, even if no individual industry becomes more productive. Productivity at a national level is therefore a mix of firms improving and the economy’s industrial mix changing.
Does weak productivity affect inflation and interest rates?
It can, indirectly. When pay rises faster than output per hour, firms’ unit labour costs go up, and some of that may be passed on in prices. Central banks watch this because it bears on how much inflationary pressure is coming from the labour market. Our explainer on why Bank of England rate decisions ripple through the economy describes the wider transmission channels.
Productivity also feeds into the economy’s supply capacity. If the amount the economy can sustainably produce grows slowly, demand may run into limits sooner, which affects how policymakers judge how much spare capacity exists. Because these judgements rest on noisy statistics, they carry real uncertainty.
What should readers watch when productivity headlines appear?
A few habits make the numbers easier to interpret:
- Check the measure. Output per hour and output per worker can tell different stories in the same quarter.
- Check the source. As the latest release shows, RTI-based and LFS-based results can differ in sign, not just size.
- Look at the year-on-year trend. Quarterly productivity figures are volatile and are often revised. The ONS flagged an operational issue in LFS collection in May 2026 that it said had a small effect on the RTI-based result: its best estimate was that growth would have been 0.6% rather than 0.7% without it.
- Remember the output side is revised too. GDP estimates change as more data arrives, which feeds straight into productivity. Our guide to how UK GDP is measured and revised explains why.
Reading slow productivity growth with caution
Productivity data are among the hardest official statistics to pin down, because they combine an output estimate with a labour input estimate, each with its own uncertainties. The current release shows that clearly: the same economy can look as if it is gaining ground or standing still depending on how workers and hours are counted.
What is clearer is the broad direction. The ONS describes growth as subdued by the standards of the years before 2008. Whether that changes depends on investment, skills, technology adoption and the mix of industries, none of which can be read from one quarter’s figure. The next release, due on 17 November 2026, will show whether the gap between the two sources narrows.
This article is general information, not personalised financial advice. Rules, rates and statistics change.
Sources: Office for National Statistics
