MNI INTERVIEW: UK Labour Force Survey Now Improved

Work on extending the coverage of the UK’s much-criticised Labour Force Survey is almost complete, making it more coherent with other data such as payrolls, though a key goal is to transition to a digital-based Transformed Labour Force Survey in the autumn, a senior statistics official told MNI.

Response rates for the LFS, which fell to as low as 17% as the survey lost its accreditation as a national statistic in 2023, raising questions about the Bank of England's ability to properly assess the labour market, have now been boosted as much as is possible within the current format, said James Benford, director general for economic, social and environmental statistics at the Office for National Statistics.

"We'll try and get a few more on the achieved response level, but I don't think we go much beyond the pre-Covid point ... there's a number of things that hold it back. It's too long, it's not digital," Benford, who was formerly the Bank of England's chief data officer, said in an interview.

The ONS assigned additional resources to the LFS from 2024, including providing bigger vouchers for respondents, he said. (See MNI INTERVIEW: UK Employment Law Tightens As Jobs Data Ease)

MORE RESOURCES

"We grew heads in the interviewing community by around a sixth last year, and that has basically lifted the achieved sample size ... to get back to close to the pre-Covid point, and we got there toward the end of last year,” Benford said.

"Since then, the outputs of that survey have been much more coherent with other indicators in line in terms of the level of employment, for example," Benford said, though he acknowledged that there was a temporary period while the LFS data was being improved when it actually seemed less in sync with other data.

"It confused things for a while ... as we improved the sample size for the Labour Force Survey, we found more employed people, and that brought the level of that survey more in line with the payrolls data," he said.

"But that was during a period when the payroll data was going down. So if you're looking at changes, it was confusing, because the LFS was going up, because it was becoming more accurate.  But with those improvements achieved, we're in a good place now," he added.

A second track to improve the LFS, by using payroll data to improve its weights and tackle non-response bias, has been less noticed, according to Benford.

"We are able to say for each household in the LFS survey, do they have payroll employees in the tax data?  And what that could offer, for example, is a more accurate way to calibrate the weights we use to draw conclusions from the survey," he said.

"A question for a while ... when the response rates in the Labour Force Survey were very low, has been … was there a bigger non-response bias in that survey? And this linkage work is going to allow us to explore how that bias has evolved over time and consider if further adjustments may be required.”

TRANSFORMED LFS

The LFS survey will eventually be superseded by the digital-based Transformed Labour Force Survey, but Benford did not give a firm timetable for this transition, saying that the aim was to transtiion to the TLFS it "in the autumn,” but that the decision on when to go ahead will be “data-led."

Benford said that data users need to be aware of the high cost of producing both the TFLS and the LFS as it costs GBP30 million to double run the surveys. 

The TLFS questionnaire has been streamlined, so that a household can do it in 15 minutes and an individual in seven minutes, he said, adding "that's up there with the best time to complete in the world.”