Reconciliation of estimates of jobs: September 2015

This report compares the latest Workforce Jobs (WFJ) estimates with the equivalent estimates of jobs from the Labour Force Survey (LFS). This is produced every quarter, when the latest WFJ estimates are released.

The concept of employment (measured by the LFS as the number of people in work) differs from the concept of jobs, since a person can have more than one job, and some jobs may be shared by more than one person. The LFS, which collects information mainly from residents of private households, is the preferred source of statistics on employment.

The LFS can also be used to produce estimates of the total number of jobs in the UK, by adding together the headline employment figures (which are equivalent to main jobs) and those for workers with a second job. The WFJ series, which is compiled mainly from surveys of businesses, is the preferred source of statistics on jobs by industry, since it provides a more reliable industry breakdown than the LFS.

A spreadsheet containing Labour Force Survey and Workforce Jobs reconciliation estimates is available on the ONS website at data table X03 (116 Kb Excel sheet).

The LFS estimate of total UK jobs for the May to July 2015 three month period is calculated by adding together the LFS figures for total employment (31.095 million) and workers with second jobs (1.188 million). On comparing this LFS UK jobs estimate (32.283 million) with the corresponding WFJ figure for June 2015, (33.694 million) the LFS total jobs estimate is lower than the WFJ figure by 1.411 million (4.4 per cent).

Figure 1 illustrates this comparison over time. These estimates have not been adjusted for factors causing differences between the two sources because many of these factors cannot be measured on a quarterly basis. Over the latest comparable quarterly periods, the LFS series shows a quarterly increase of 17,000 jobs (0.1 per cent) and the WFJ series shows an increase of 102,000 (0.3 per cent). On an annual basis the LFS series shows an increase of 400,000 (1.3 per cent) and the WFJ series shows an increase of 330,000 (1.0 per cent).

The 2006 National Statistics Quality Review of Employment and Jobs Statistics (4.46 Mb Pdf) identified about 30 reasons why the LFS and WFJ estimates of jobs can differ from each other. Some of these factors can be quantified approximately using information from the LFS and other sources, while others are much more difficult to measure. The measurable factors causing differences between the LFS and WFJ figures are included in a downloadable spreadsheet within the ‘download chart’ option of this report.

Methodological changes to the reconciliation have been made for the December 2014 release implemented for the September 2014 results. These changes follow on from changes made in the April 2013 release.

1. Temporary foreign workers: previously the time lag between the latest International Passenger Survey (IPS) data publication and the quarterly adjustment was filled by applying the annual growth in National Insurance Number registrations (NINo’s) to the most recent IPS figure. The new method does not use NINo registrations because they are no longer regarded as a reliable forward indicator of changes in the number of temporary foreign workers. The latest IPS estimate for short term migrants in employment will be carried forward until new IPS data is available.

2. Workers living in communal establishments were adjusted for using pilot survey data from 2000. 2011 Census data has now replaced this pilot survey source.

3. The adjustment for response bias on the LFS has been removed from the reconciliation.

4. Estimation of the over counting of self-employed jobs now incorporates data relating to second jobs as well as main jobs.

5. WFJ has been benchmarked to the 2013 Business Register and Employment Survey (BRES).

6. All LFS estimates have been reweighted to population estimates based on the 2011 Census.

Figure 2 shows the two jobs series adjusted to take into account the measurable factors causing differences between the LFS and WFJ statistics. Once these factors have been taken into consideration, the adjusted LFS estimate of total UK jobs is lower than the adjusted WFJ estimate, by 786,000 (2.4 per cent).

The difference between the adjusted LFS and WFJ estimates (786,000) is beyond the likely bounds of the sampling variability of the difference. The approximate sampling variability (95% confidence interval) is roughly ± 300,000 to ± 400,000. However, it should be noted that the adjustments are themselves subject to a margin of uncertainty, and there are other factors causing differences between the two sources which have not been adjusted for. There are about 20 additional factors that could explain the remaining difference between the LFS and WFJ estimates. As well as sampling variability, they include, for example, timing effects. The LFS estimates are averages for 3-month periods, whereas business surveys measure the number of jobs on a particular day.