Public sector employment, UK: September 2016

This quarter’s release

This quarter revised data for the NHS in England have been included in our public sector employment statistics from March 1999 onwards, as a result there have been revisions to the series that include this data. These revisions result from improvements made to their data collection and the consultation held in 2015. These revisions bring the figures used for our PSE statistics in line with the figures published by NHS Digital.

Private registered providers (PRPs) of social housing (referred to as housing associations in this bulletin and related datasets) in England were reclassified from the private to public sector on 30 October 2015. In this release we have implemented this classification decision and therefore the series have been revised from September 2008 onwards. More information can be found in the Classification announcement: “Private registered providers” of social housing in England report.

Relevance to users

The PSE estimates and data produced for the quarterly publication are used across government and feed into a number of wider publications and outputs. Some government departments use the total figures to facilitate policy making, whereas others use specific components of the data collection. The main users are as follows:

• Cabinet Office

• HM Treasury

• Scottish Government

• Department for Communities and Local Government (DCLG)

• Local Government Association (LGA)

Revisions

Public sector employment statistics have previously been published for all periods from 1999 up to and including June 2016. In line with the published revisions policy for public sector employment statistics, the statistics have been revised, to take account of late information from respondents.

Tables 1R to 5R in the public sector employment dataset illustrate the size of the revisions in each category.

Concepts and definitions

The public sector comprises central government, local government and public corporations as defined for the UK National Accounts. We publish the Public sector classification guide monthly, and provide information on the classification of organisations and institutions in the National Accounts.

The public sector employment estimates relate to the number of people employed according to returns from relevant organisations, but they include a number of workers with a second job in the public sector whose main job is in the private sector or in a separate public sector organisation. The private sector estimate, which is obtained by taking the difference between the Labour Force Survey estimate of people employed in the whole economy and the public sector total, will therefore tend to be correspondingly understated by a small percentage.

Headcount estimates are based on the number of employees with an employment contract who are being paid by the organisation. Employees can be permanent, on a fixed-term contract or employed on a casual basis. Self-employed, contract workers and agency workers are excluded.

Permanent employees, as defined in the public sector employment dataset Tables 8 and 10, are employees with a contract with no agreed expiry date or a fixed-term contract of more than 12 months. Temporary or casual employees are those with a fixed-term contract of 12 months or less or employed on a casual basis.

As well as the headcount estimates, estimates have also been produced for the number of employees in full-time equivalents (FTE) back to 1999. This is based on converting part-time employees’ hours into a full-time employees’ equivalent and provides a better indication of total labour input than a simple headcount.

Central government includes all administrative departments of government and other central agencies and non-departmental public bodies. As such it is wider than the Civil Service. This sector also includes HM Forces and the National Health Service (NHS). Within education, academies and free schools are classified to central government. It also includes the British Transport Police in England and Wales and, from June 2013, the Police Service of Scotland.

Local government covers those types of public administration that only cover a locality and any bodies controlled and mainly financed by them. The sub-sector includes all areas of administrative authorities including parish councils, though these units are not covered by the current estimates for local authorities. It includes police forces and their civilian staff for England and Wales, excluding British Transport Police. Until June 2013 it included the Police Service of Scotland. All functions of local authorities are classified to the sub-sector, although trading activities that produce market output (for example, housing and municipally owned markets) are regarded as quasi-corporations and appear under public corporations. Local education authorities are part of local government, as are voluntary aided schools, county schools and, from September 1999, foundation schools (formerly grant-maintained). Public corporations are companies or quasi-corporations controlled by government, for example London Underground Ltd. These companies receive more than half their income from sales of goods or services into the market place.

The estimates of Civil Service employees count all home Civil Service employees. Civil Service employees can be classified to central government or public corporations. Examples of public corporations include the UK Intellectual Property Office and the Driver and Vehicle Standards Agency. Civil Service estimates exclude the Northern Ireland Civil Service and other Crown servants. Employees in these groups are included in estimates of central government employment.

Accuracy

Response rates:

PSE statistics are compiled from a range of sources. The primary source is the Quarterly Public Sector Employment Survey (QPSES). The QPSES comprises 3 separate data collections; the home Civil Service, Local Authorities in England and Wales, and Great Britain public corporations and non-departmental public bodies (NDPBs). Returned questionnaires go through a series of automated validation tests to check for completeness and consistency and to identify any significant movements compared with the previous period reported (and the same period the previous year). The automated checks are followed up with respondents where errors are detected or further explanation is required. The target is to clear 95% of test failures prior to processing results. Our targets for response to Local Authorities and Public Bodies QPSES before the results are compiled are 90% (number of respondents) and 90% (of total employment). In addition, each survey has a list of critical respondents (usually those with the largest employment) for which special efforts are made to achieve 100% response and clearance of test failures. For Civil Service QPSES the targets for response before the results are compiled are 100% (number of respondents) and 100% (of total employment).

Data for non-responders are imputed based on previous returns and known annual changes in seasonality. It is extremely rare for a local authority, public body or Civil Service department to non-respond for 2 consecutive quarters. The data collection is statutory for local authorities and public bodies (Statistics of Trade Act 1947) and positive action is taken to address non-response issues as and when they occur.

So that estimates of total public sector employment can be made, it is necessary for further information to be gathered from external sources.

We also produce regional estimates of PSE based on returns from public sector organisations (Table 6). These supersede those produced using the Labour Force Survey (LFS) which previously had been used in conjunction with national PSE estimates to produce estimates by region.

Coherence

Estimates of public sector employment are based partly on projections for some sources. As part of the development programme to improve the quality of public sector employment estimates, public sector organisations are working towards the production of timely quarterly estimates. Until this development programme is completed, there remains a requirement to include estimates for certain sources:

i. Police workforce estimates (including Special Constables) for England and Wales are published every 6 months (for 2 quarters ) by the Home Office. These quarterly estimates are therefore based on projections, and may be subject to revisions when new data are published.

ii. NHS workforce statistics for England are derived from a pay system which covers all but 2 English NHS organisations. This produces very good estimates of staff numbers. Figures for the 2 other organisations are estimated based on quarterly NHS Workforce figures.

iii. Estimates of employment in housing associations in England are derived using data provided by the Homes and Communities Agency (HCA) and data from the Inter- Departmental Business Register (IDBR) . The latest period of actual data from HCA relates to March 2015, after this the estimates are based on forecasts and will be subject to revisions in the next revisions round (September 2017) when the series will be revised based on the 2016 HCA data.

All time series in the public sector employment release, except for the regional series, are seasonally adjusted to aid interpretation. As seasonal adjustment does not preserve additivity within aggregation structures, relationships that hold in the unadjusted series do not necessarily hold for the seasonally adjusted series. For example, total public sector employment equals the sum total of all public sector industry estimates before seasonal adjustment, but this is not necessarily true after seasonal adjustment.

The estimates of public sector employment in education (SIC division 85) differ from the school workforce estimates published by the Department for Education (DfE) mainly as a result of differences in coverage and data sources. DfE estimates focus on the number of full-time equivalent (FTE) teachers and support staff for England only. By comparison, our estimates are derived by allocating local authority employees to education using the IDBR and the QPSES in England and Wales. The DfE School Workforce Census school level estimates are used to estimate employment in academies in England. PSE estimates include all employees reported by local authorities as working in primary, secondary and adult education establishments including some groups who are not covered by the DfE statistics, such as adult education staff and certain categories of support staff. Employment estimates for education in Scotland and Northern Ireland are also included to give a wider UK coverage. The different coverage of the Office for National Statistics (ONS) and DfE education statistics serve the needs of different users. Those who require information on the workforce in England who are directly involved in pupils' teaching and learning should use DfE published statistics. Users should also refer to DfE published statistics to gauge trends in education employment. Those who seek data on UK public sector employment in education, in its widest sense, should use our data in this release. For further information on the differences between DfE and ONS data on education please see pages 44 to 46 of the Public Sector Employment Trends 2005 article, published in October 2005.

Our estimates for the NHS also differ from the headline figure produced by NHS Digital. Again, this reflects the wider UK coverage (NHS Digital figures are for England only) plus our exclusion of general practitioners (GPs). In accordance with National Accounts practice, we classify GPs as part of the private sector. When these factors are allowed for, ONS and NHS data can be shown to be identical.

Machinery of government changes in the period since 1 October 2015 are listed here:

Methods

Improvements to the way employment in public sector education in England is estimated were first implemented as part of the PSE, Quarter 3 2012 release. Revisions to the estimates caused by these improvements were at that time incorporated into the revised PSE series, in line with the revisions policy for public sector employment statistics. Further details of the change in method and the impact on estimates of PSE are available in Public Sector Employment Statistics - Change in Method for Estimating Employment in Education in England, published as part of the Public Sector Employment, Quarter 3 2012 release.

Estimates of employment in housing associations in England are derived using data provided by the Homes and Communities Agency (HCA) and data from the IDBR. HCA provide annual full-time equivalent (FTE) data for large housing associations with more than 1,000 properties. The large housing associations have been matched to the IDBR, the IDBR organisation structures are used to calculate industry and regional estimates. The ratio between headcount and FTE for the other public corporations is used to gross the FTE estimates up to give headcount estimates. A quarterly path has been derived from the annual data.

The estimates for large housing associations, with more than 1,000 properties, are supplemented by estimates for the remaining small housing associations to create a total estimate. The estimates for small housing associations were calculated from the available IDBR matches. The small housing associations were stratified and estimates were created based on the known employment distributions within each stratum. To create a time series this estimate has been modelled back based on the known movements of the large housing associations.

The HCA data are only available annually and therefore some forecasting is required to produce timely employment estimates. The latest period of actual data from HCA relates to March 2015, after this the estimates are based on forecasts and will be subject to revisions in the next revisions round (September 2017) when the series will be revised based on the 2016 HCA data.

Reclassifications

In recent years, the public and private sector employment series have been affected by a number of major reclassifications, where bodies employing large numbers of people have moved between the public and private sectors. These major reclassifications are as follows.

Further education corporations and sixth form college corporations in England are included in the private sector from June 2012 but in the public sector for earlier time periods. More information on this decision can be found in the Reclassification of Further Education Corporations and Sixth Form Colleges in England article, published on 31 May 2012.

Royal Mail plc is included in the private sector from December 2013 but in the public sector for earlier time periods.

Lloyds Banking Group plc is included in the public sector from December 2008 to December 2013 but in the private sector for earlier and subsequent periods.

Royal Bank of Scotland plc is included in the public sector from December 2008 but in the private sector for earlier time periods.

Network Rail is included in the private sector before December 2002. From December 2002 onwards it is included in the public sector (except for the period from June 2003 to March 2004, when it is included in the private sector). More information can be found in the Classification of Network Rail under European System of Accounts 2010, published in December 2013.

Northern Rock is included in the public sector from December 2007 until December 2011 but in the private sector for earlier and later time periods.

Bradford and Bingley is included in the public sector from September 2008 but in the private sector for earlier time periods.

Welsh further education colleges were reclassified to the private sector on 27 January 2015. They are included in the private sector from March 2015 but in the public sector for earlier time periods.

Private registered providers (PRPs) of social housing (referred to as housing associations in this bulletin and related datasets) in England were reclassified from the private sector to the public sector on 30 October 2015. In this release we have implemented this classification decision and therefore the series have been revised from September 2008 onwards. More information can be found in the Classification announcement: “Private registered providers” of social housing in England report, published in October 2015.

Subsequently the date this classification applies from was amended in the September 2016 public sector classification guide update and now applies from 24 July 1996. In addition, registered providers of social housing in the devolved administrations were reclassified from the private to public sector at this time. Further work will be required to determine if it is possible to implement the full decision for England and how this should be implemented for the devolved administrations. More information on the classification decision for registered providers of social housing in the devolved administrations can be found in the Statistical classification of registered providers of social housing in Scotland, Wales and Northern Ireland: September 2016 report.

Comparisons of public and private sector employment over time are complicated by a number of changes to the composition of these sectors over this period with several large employers moving between the public and private sectors. We therefore publish estimates of public and private sector employment, excluding the effects of major reclassifications, alongside estimates of total public and private sector employment in Tables 5, 6a and 7a of the PSE release.

On 13 October 2010, we announced the reclassification of further education colleges and sixth form college corporations to the public sector. As part of the December 2010 publication, we took on employment estimates for further education colleges back to 1993 or their inception if later.

On 31 May 2012, we announced the reclassification of English further education colleges and sixth form college corporations to the private sector, as non-profit institutions serving households (NPISH), effective from 1 April 2012. As such, employment estimates for English further education colleges and sixth form college corporations are included in our estimates of public sector employment from 1993 or their inception if later, up to and including March 2012.

English further education colleges and English sixth form college corporations estimates of employment are not included in public sector employment estimates from June 2012 onwards.

Publication policy

The complete run of public sector employment data in the tables of this statistical bulletin is also available to view and download in other electronic formats free of charge using our Time Series Data website service. You can download the complete Public sector employment time series in a choice of zipped formats, or view and download their own selections of individual series.

A list of the job titles of those given pre-release access to the contents of this statistical bulletin is available on our website.

Details of the policy governing the release of new data are available by visiting the UK Statistics Authority website.

The UK Statistics Authority has designated these statistics as National Statistics, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Official Statistics.

Designation can be broadly interpreted to mean that the statistics:

• meet identified user needs

• are well explained and readily accessible

• are produced according to sound methods

• are managed impartially and objectively in the public interest

Once statistics have been designated as National Statistics it is a statutory requirement that the Code of Practice shall continue to be observed.

As stated earlier in the bulletin, the employment figures provided are point-in-time estimates and for this reason, we have introduced a new naming convention for the releases, whereby the latest month of measurement is highlighted rather than the quarter.