1. Abstract

Workforce Jobs (WFJ) is a quarterly measure of the number of jobs in the UK and is the preferred measure of the change in jobs by industry. It is a compound source that draws on a range of employer surveys, household surveys and administrative sources. WFJ is the sum of employee jobs (EJ) measured primarily by employer surveys (predominantly the Short-Term Employment Surveys (STES) and the Quarterly Public Sector Employment Survey (QPSES)), self-employment jobs (SEJ) from the Labour Force Survey (LFS), and government-supported trainees (GST) and Her Majesty’s Forces (HMF) from administrative sources (see Annex A (England GST data from 2006 onwards now from LFS)). A variety of outputs by industry, region, gender and full- or part-time status are produced for a range of publications and users.

This article explains, in detail, the revisions that have been made to the WFJ series since June 2016 as a result of:

  • benchmarking to the latest estimates from the annual Business Register and Employment Survey (BRES); these revisions will go back to 2012

  • changes to seasonal parameters following a seasonal adjustment review; these revisions will go back to 1981

  • revisions to Public Sector Employment (PSE) (impact on WFJ begins in 1996 Q1)

  • revisions to STES for periods 2015 Q2 onwards, due to updates in data supplied or late data

  • re-weighting of series sourced from the LFS back to 2012 Q2,

  • refinements in methodology used to revise series back to 1959 Q2 due to inclusion of single Pay As You Earn (PAYE)-based business into sample selection from 2016 Q1,

  • revisions back to 2006 Q2 to the GSTs series due to improvements in the source of data

Figure 1 compares the previously published and revised UK WFJ (seasonally adjusted).

Nôl i'r tabl cynnwys

2. Benchmarking Great Britain employee jobs to the Business Register and Employment Survey

Benchmarking is an annual process to align the quarterly GB EJ series to the latest estimates from the Business Register and Employment Survey (BRES). BRES is based on a sample of approximately 80,000 reporting units1, a much larger sample than STES, and so generally is considered to produce more accurate and detailed estimates of the level of employment. BRES estimates refer to Q3 of a given year. The private sector element of GB EJ series has been benchmarked to the equivalent from BRES for the periods 2014 Q3 and 2015 Q3 (the latest period available).

Notes for: Benchmarking Great Britain employee jobs to the Business Register and Employment Survey

  1. Reporting units hold the mailing address to which survey questionnaires are sent. The questionnaire can cover the enterprise as a whole, or parts of the enterprise identified by lists of local units.
Nôl i'r tabl cynnwys

3. Seasonal adjustment

Seasonal adjustment is the process of identifying and removing the seasonal components from a series to leave the underlying trend and irregular components. The revised WFJ series have undergone a seasonal adjustment review, by ONS’s Time Series Analysis branch, causing revisions back to 1981 Q3.

Nôl i'r tabl cynnwys

4. Changes to estimates of public sector employment

WFJ makes use of ONS’s official public sector employment (PSE) estimates for GB. These inputs are not benchmarked as they are the definitive measure of PSE. The data received at 2016 Q3 have led to revisions back to 1996 Q1 in WFJ. The main cause of the negative revisions back in time is revised employment data on the NHS. Further information is available in the 2016 Q3 PSE Statistical Bulletin.

Nôl i'r tabl cynnwys

5. Late or revised data

The private sector employee jobs data comes from sample surveys. Each quarter, revisions to data supplied or late data are incorporated for the previous quarter in line with the revisions policy. Revisions or late data for earlier periods (2015 Q2 onwards) have now been included.

Nôl i'r tabl cynnwys

6. Re-weighting of Labour Force Survey estimates

The LFS is used to measure SEJ. It is also the source of EJ estimates for agriculture, activities of households as employers, and construction up to 2010 Q3. The LFS was re-weighted in May 2016 with revised series being produced back to 2012. Revisions have been taken on for all domains.

Nôl i'r tabl cynnwys

7. Improvements in coverage of business survey population

Revisions were made to the workforce jobs series back to 1959 in the 2016 Q1 release. This was in order to maintain a comparable back series in response to an increase in business survey coverage to include solely Pay As You Earn-based businesses from January 2016.

Subsequently there has been a refinement in the methodology used to calculate this revision. Therefore the figures between 1959 Q2 and 2015 Q4 have been re-adjusted, causing downward revisions to all periods ranging from 83,000 (0.28%) in 2015 Q4 to 52,000 (0.22%) in 1959 Q2.

Nôl i'r tabl cynnwys

8. Government-supported trainees

There has been an improvement in the methodology used to estimate government supported trainees (GSTs) in England. Previously the entire data series was supplied by the Department for Business, Energy and Industrial Strategy (BEIS), but for all periods from 2006 Q2 onwards the source has been switched to the Labour Force Survey (LFS). The new WFJ GST series from 2006 Q2 uses an adjusted version of the GST figures published by LFS, which takes into account respondents that do not adhere to the WFJ definition of a GST. The series from 2006 Q2 onwards are also based on the pattern from LFS GST respondents to produce regional and gender breakdowns and, whereas previously these were supplied through the data from BEIS.

Nôl i'r tabl cynnwys

10 .Annex A – Workforce jobs sources

Nôl i'r tabl cynnwys

11 .Annex B – Revision tables, Workforce Jobs by Component

Manylion cyswllt ar gyfer y Methodoleg

Mark Williams
Ffôn: +44 (0)1633 456776