UK’s productivity puzzle solved by statistical overhaul

Last month the UK’s labour productivity figures doubled. This wasn’t an example of workers becoming more diligent and industrious; it resulted from a modernising in the way the Office for National Statistics (ONS) gathered its figures.

The UK has long been an international outlier when it comes to productivity, with the ONS reporting that between 2009 and 2019 productivity annual growth was 0.7%. The new figures for the same period are nearly double that at 1.3%.

This overhaul of how it is measured follows years of concern that the country’s labour market data has been inaccurate.

Labour productivity, commonly measured as output per hour worked, is one of the most important indicators of economic performance because it shows how efficiently labour is being used. Before the 2008 financial crisis, UK productivity grew by around 2% annually, but officially it collapsed to just 0.7% a year between 2009 and 2019.

This ‘productivity puzzle’, it turns out, was due to flaws in statistical measurement.

After years of research and following recommendations made by the Organisation for Economic Co-operation and Development (OECD) as far back as 2018, the ONS accepted that some of the problem lay in how labour input was being measured.

While Gross Domestic Product (GDP) and Gross Value Added (GVA) figures remain unchanged, the ONS estimates of hours worked have been improved significantly. It seems that not taking account of holidays, for example, has proved critical to the miscalculations.

What was the problem?

A key issue stemmed from the long-running Labour Force Survey (LFS), which asked individuals how many hours they had worked in the previous week. The weaknesses in this approach were that employees often struggled to recall their actual hours and tended to report their usual or contracted hours instead.

The problem was compounded when one household member answered on behalf of others.

Falling survey response rates created further distortions. During periods such as July and August, when many workers were on holiday, missing responses were often assumed to reflect normal working patterns rather than reduced hours because of annual leave. As response rates declined, particularly in recent years, this has led to a growing overstatement of hours worked across the economy.

The consequence was significant. If total output is divided by an inflated number of hours, productivity appears weaker than it really is. The ONS believes this measurement issue partly explains why UK productivity growth looked lower than that of comparable countries.

New approach

To address these shortcomings, the ONS has developed a new ‘component approach’ that combines information from several datasets rather than relying heavily on a single survey.

Under the revised methodology, estimates of jobs are drawn from the Workforce Jobs series, which is based on employer data and better captures those working in the UK economy. These figures are supplemented with HMRC’s Real-Time Information (RTI) payroll records, which provide near real-time data on employees on company payrolls.

The approach to measuring hours worked has also changed substantially. Instead of asking workers to recall actual hours worked in a particular week, the ONS now uses usual working hours from the Annual Survey of Hours and Earnings (ASHE). It then adjusts these figures using information on annual leave, sickness absence, unpaid overtime and other factors that affect actual working time. Additional employment information comes from the Business Register and Employment Survey (BRES) and quarterly business surveys.

This combination of administrative records and business surveys follows international best practice and creates a more robust picture of labour input.

The new analysis suggests that average hours worked have fallen more sharply since the financial crisis than previously thought. As a result, productivity growth per hour was stronger than earlier estimates indicated.

A productivity slowdown remains, but the scale of the puzzle has effectively been cut in half.

For businesses, the takeaway highlights the importance of accurate data. Reliable statistics provide a clearer understanding of economic performance and offer a stronger foundation for decisions on investment, innovation and growth.