Major Changes to the 'Official' Definition of Unemployment and Unemployment Series
In January 2003, the Office for National Statistics (ONS) announced that it had changed the 'Official' source of unemployment data from figures based on the Claimant Count to those derived from the Labour Force Survey (LFS). Claimant Count data will continue to be published monthly by the ONS to provide additional information about the labour market, but these will not be presented as 'an alternative' to the 'Official' LFS measure of UK unemployment.
The figures provided in the above Summary Table are sourced from the Claimant Count data and therefore no longer represent 'Official' Unemployment Figures, but simply a measure of those people who are claiming unemployment-related benefits (i.e. the Claimant Count). The latest 'Official' unemployment figures for the UK and North West are provided in paragraphs below.
Figures based on the Labour Force Survey, use the International Labour Organisation (ILO) definition of unemployment, which is the only internationally comparable measure of unemployment in the UK. This is now the 'Official' measure of UK unemployment that will be reported via the press, TV, radio and other media. The Labour Force Survey definition of unemployment is broader than that used previously via the Claimant Count and as a result the 'Official' numbers and rates of unemployment reported in the media will be greater than reported under the previous Claimant Count definition. The LFS definition of unemployment is given at the end of this document.
Please note that the ONS also changed the denominator used to calculate Claimant Count proportions (previously called rates) as from December 2002 onwards (see explanatory paragraphs below). The proportions reported as from December 2002 on the Summary Sheets are not comparable with the previous rates reported on the Summary Sheets prior to December 2002.
Why Does The Above Summary Sheet Table Not Contain 'Official' Labour Force Survey Figures?
Labour Force Survey v the Claimant Count - Availability of Data at the Local Level
Unemployment figures sourced via the Labour Force Survey for smaller areas, such as Districts and Wards, are often based on survey samples of insufficient size to provide reliable figures. On such occasions, the figures are suppressed and not released for publication. This results in gaps in local Labour Market information. Figures sourced via the Claimant Count, however, although no longer 'Official', are not sample based and continue provide a fairly consistent, timely and useful measure of the Labour Market at the local level. In light of the availability of the Claimant Count data at the local level, Lancashire County Council will continue to publish figures sourced via the Claimant Count to maintain a local labour market perspective.
Please note the ONS published the mid-year population estimate (MYE) for 2003 on 9th September 2004, along with revised mid-year population estimates for 2001 and 2002. These estimates incorporate revisons in relation to the underestimation of the population in Manchester (and 14 other local authority areas) encountered via the 2001 census. The revisions have been incorporated into LFS unemployment estimates as from the three months ending in August 2004 (and also from 1992 to date - for data published via the Labour Market First Release since October 2004). Previously, the 2001 census of population had underestimated the population in Manchester by 20,000 in 2001 and by 22,000 in 2002. Overall, the effect on the figures has been small.
Release of 'Official' LFS Annual Unemployment Figures for Lancashire County, Districts And Unitary Authorities
Revised Annual Local Area LFS data for 2001 were made available on 17th March 2004. These data have been re-weighted to make them consistent with the population estimates published in February and March 2003, which were revised in light of the 2001 Census results. The local area LFS data used in the local area profiles are also updated. The 2002 local area Annual data are scheduled to be published on 28th April 2004. Revised Annual data for 1996, 1999 and 2000 are scheduled to be made available by 7th April 2004. Revised Annual data for 1997 and 1998 will be published about August/September 2004.
Please note that LFS Annual unemployment rates at the District and Unitary Authority level are often based on survey samples of insufficient size to provide reliable figures. On such occasions, the figures are suppressed and not released for publication.
Notes on the Claimant Count Figures Contained in the above Summary Sheet Table
The data in the above Summary Table no longer represents 'Official' Unemployment Figures, but simply a measure of those people who are claiming unemployment-related benefits (i.e. the Claimant Count). Lancashire County Council will continue to publish 'unadjusted' Claimant Count figures (as opposed to seasonally adjusted data). The Male, Female and Total Claimant Count numbers reported on the Summary Sheets as from December 2002 are comparable with numbers on Summary Sheets up to and including November 2002, however, the new proportions (previously labelled rates until November 2002) contained on the Summary Sheets as from December 2002 are no longer comparable with those prior to December 2002 (see paragraphs immediately below).
The ONS/NOMIS introduced 2003 Census Area Statistic (CAS) Wards to collate Claimant Count Numbers at the Ward level for data relating to February 2004 onwards. Prior to February 2004, the monthly Claimant Count Numbers related to 1991 frozen wards (which were introduced in approximately 1996). The ONS/NOMIS will continue to build up District level Claimant Count statistics (as well as those for Travel to Work Areas, Government Office Regions and Constituencies) from 1991 frozen ward data until the ONS/NOMIS system is upgraded during 2005/2006 to use 2001 Census Output Areas. In a very few instances, this may result in some minor discrepancies with the figures at the district level and above (as the sum of the Claimant Count derived via the 2003 CAS Wards and their Proportions may differ marginally from those derived via the 1991 Frozen Wards).
Withdrawal of Claimant Count Workplace Based Rates / Introduction of Claimant Count Proportions of the Resident Working Age Population
In January 2003, the ONS announced that as from December 2002 it would be publishing local area Claimant Count levels as a proportion of the resident working age population (females 16-59, males 16-64) as opposed to the previous rate that used a workplace-based denominator. The new residence-based measure is not distorted by the effects of commuting and was a main reason for implementing the changes to the unemployment data.
The previous workplace-based Claimant Count rates used until November 2002 for areas smaller than regions (i.e. Districts, Unitary Authorities, Counties, TTWAs and Parliamentary Constituencies) have been withdrawn from National Statistics and no longer appear.
The new percentage figures are referred to as 'proportions' to distinguish them from workplace-based 'rates'. The new proportions use resident working age populations derived from the mid-2001 population estimates, which are compatible with the 2001 census, published by the ONS on 10th October 2002. The new proportions are not comparable with rates on Summary Sheets dated prior to December 2002.
Please note that NOMIS has not previously produced unemployment rates or proportions at the ward level. It is expected that 'Official' proportions of the Resident Working Age Population for wards will become available via NOMIS as from May 2004.
Incompatibility of New Series with Previous Ward and District Residence-Based Unemployment Rates Produced by Lancashire County Council
In order to provide unemployment rates that were not distorted by the effects of commuting (that was evident with the ONS Workplace rates) Lancashire County Council's Research and Intelligence Team also produced estimated unemployment rates until November 2002 for those living in an area at the District and Ward levels using a denominator based upon the 1996 workforce estimate.
The new residence based proportions at the district and ward level are however not comparable with district rates produced prior to December 2002 based on either the previous ONS workplace-based denominators or the Lancashire County Council 1996 residence-based workforce estimates, as the methodology used to calculate the denominators differ.
Claimant Count Proportions for Travel To Work Areas
Proportions for Travel To Work Areas are now available via the NOMIS website using working age population denominators derived from the 2001 Census of Population output areas, which were published on the 30th June 2003. These have been adjusted to make them consistent with 2001 mid-year population estimates.
Back Dated Series for the New Residence-Based Claimant Count Proportions (Previously Rates)
New residence-based Claimant Count proportions are available for District and Unitary Authority areas from January 1992 onwards.
An Explanation of the Distortions Experienced with Previous Workplace-Based Claimant Count Rates and Comparison Against the New Claimant Count Residence-Based Proportions.
Previous unemployment rates derived from workplace-based denominators can be distorted by commuting patterns. Such rates can therefore be misleading as indicators of local labour market conditions. The new figures based on the proportions of people of working age living in an area who are claiming jobseeker's allowance do not suffer from this shortcoming.
It has been estimated that on average the new residence-based proportions should be around a fifth smaller than the previous workplace-based rates, owing to the new resident population of working age denominator usually being larger than the workplace denominator used previously. Burnley and Ribble Valley would appear to reflect this generality, although the differences may change over the year as seasonal changes are reflected in employment patterns. There are however variations to this norm. For example using November 2002 Claimant Count data, the rate on the former basis in Preston was 2.4%. Expressed as a Proportion of the residence-based working age population, however, the November figure for Preston is 2.6%, marginally higher. This reflects a situation where the workplace denominator (the number of jobs in Preston and those claiming unemployment benefit) is larger than the resident population of working age. A similar situation is evident for Fylde District.
For areas that are not self-contained labour markets, the previous workplace-based rates tended to overstate the incidence of claimant unemployment where there were only a few jobs in the area but where many residents commute out of the area. Within Lancashire this would seem to be particularly apparent for Wyre, Pendle and Chorley and to a lesser degree for Hyndburn, West Lancashire, Lancaster, Rossendale and South Ribble. In contrast, workplace-based rates would tend to underestimate the problems among residents in areas where there are a large number of people commuting in from other areas. Within Lancashire, as described above, this would appear to be the case for Preston and Fylde districts.
What Explains the Differences? Why are the New Claimant Count Proportions Larger than the Old Claimant Count Percentage Rates in Some Cases?
The differences are mainly explained by the fact that the number of jobs in an area may be different from the resident population of working age as a result of:
- Commuting patterns, which can work either way, the new Proportions being higher than the old rates for some areas.
- Different Proportions of the resident population who are working or seeking work (i.e. differences in economic activity rates)
How are Workplace-based Claimant Count Rates Calculated?
Workplace-based Claimant Count rates express the number of claimants of unemployment-related benefits as a percentage of the workplace denominator (workforce jobs in the area plus the number within the claimant count).
How are Workforce Jobs Derived?
Workforce jobs are obtained from the corresponding mid-year estimate (or the latest available mid-year estimate). The number of workforce jobs is mainly collected through postal employer surveys and comprises employee jobs, self employed jobs, HM armed forces and government supported trainees. The previous Claimant Count rates were mainly workplace-based but some components of the denominators (the Claimant Count and the self employed) are residence based.
Availability of Denominators for Claimant Count Workplace-Based Rates
For continuity, workplace-based rates will remain available for regions and countries. These figures will continue to be updated annually. Figures used to calculate sub-regional workplace-based rates will remain accessible in the NOMIS data set 'Claimant Count denominators - historical workforce series' - but will no longer be updated. Users may use these figures to calculate sub-regional workplace based rates themselves, however, they will not be supported by the Office for National Statistics.
The New 'Official' Labour Force Survey (LFS) Definition of Unemployment
The new 'Official' source of unemployment figures, sourced via the Labour Force Survey is a broader internationally recognised measure of unemployment. The definition includes people who are aged 16 and over and:
- are without a job, want a job, have actively sought work in the last four weeks and are available to start work in the next two weeks, or
- are out of work, have found a job and are waiting to start it in the next two weeks.
Please note that the 'Official' unemployment figures sourced from the Labour Force Survey will be re-labelled 'Unemployment' rather than 'ILO Unemployment'. This will emphasise that the LFS figures provide the 'Official' and only internationally comparable, measure of UK unemployment. The LFS methodology uses the economically active workforce estimate as the denominator to calculate the unemployment rate.
This page was compiled by Paul Ayre.
All enquiries from the press should be sent to Corporate.Communications@lancashire.gov.uk.
Any other questions about the content of this page may be sent to EconInfo@lancashire.gov.uk.
For all enquiries about the county council's services, contact the Customer Service Centre on 0845 0530000 (01772 530000) or at Enquiries@css.lancscc.gov.uk.