Cabinet Office to scrutinise failings of UK’s official data provider ONS

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

One of the government’s most senior civil servants has been tasked with getting to the bottom of the problems with data produced by the Office for National Statistics, as concern grows in Westminster about the reliability of the numbers used to steer the UK economy. 

Pat McFadden, the Cabinet Office minister, has asked his top official Cat Little to look into the performance of the ONS amid fears that chancellor Rachel Reeves has been left flying blind as she draws up her plans for the economy.

“Pat thinks it’s an obvious problem,” said one colleague of the minister. “He’s raised it with the permanent secretary [Little] and has said it’s crucial we have data on which we can rely.”

Little and her colleagues have been asked to provide options for possible government action after a series of revelations about errors in official economic indicators, according to people with knowledge of her briefing. Among the ideas on the table is an independent review, although no decisions have been made. 

Last week Dame Meg Hillier, the chair of the influential Treasury select committee, wrote to Sir Ian Diamond, the national statistician, to highlight “troubling errors and delays” in some sets of statistics. 

One ally of Reeves said: “There’s frustration that we are taking economic decisions when the data is wrong. It’s a bit of a problem, to put it mildly.”

The ONS has been grappling with issues with some of the most market-sensitive data it produces, which is making it harder for UK policymakers to steer the economy. Concern has centred on the breakdown of the key Labour Force Survey, which underpins jobs data, but the ONS has found problems in other parts of its output. 

The data cruncher has suspended publication of two producer price indices that help calculate GDP, and has been criticised for “jumbled economic reasoning” by the Institute for Fiscal Studies over the way it revalued pension wealth. 

The Bank of England’s Monetary Policy Committee this month underlined the need for “high quality and reliable official data across the full range of economic and labour market statistics”, after governor Andrew Bailey bemoaned the quality of the jobs numbers. 

The Office for Statistics Regulation, an arm of the UK Statistics Authority, which oversees the country’s statistical system, is planning to publish an interim report in April on a systemic review of economic statistics it launched last year. 

⁠An official close to McFadden said he was concerned about the issues at the ONS. “It’s essential that we have reliable data and the ONS is vital to that,” the source said. 

In her letter, Hillier told the statistics chief that she was “concerned about the impact of unreliable labour market statistics and whether there are problems with other economic data”.

She added: “The latest troubling errors and delays in trade data and Producer Price Index data will widen concerns about the trustworthiness and accuracy of the economic statistics available to policymakers and other users.” 

Hillier asked for an update on the nature and causes of errors identified in the producer prices numbers, as well as the ONS’s response to the criticisms from the IFS of the wealth numbers. 

She also asked for more information about problems in the trade data after the ONS on March 13 flagged an error in the numbers dating back to 2023.

On Friday, the ONS released updates to its trade data after examining that problem. Among the issues were errors in data provided by HM Revenue & Customs relating to imports of goods, and an error in the system processing services numbers, the ONS said. 

However, the revisions to trade numbers for 2023 and 2024 did not affect wider GDP data.

The Cabinet Office declined to comment. The ONS told the FT it “remains focused on producing the highest quality statistics for the public good and is committed to the continuous improvement of our methods and approaches”.