Publication of Jobcentre Plus vacancy statistics

Publication of Jobcentre Plus vacancy statistics

Bentley, Russ

Key points

* Publication of Jobcentre Plus vacancy statistics was deferred in 2001 because of distortions in the data.

* Publication of some vacancy data is being restored by the DWP to provide an insight into Jobcentre Plus performance. The statistics will, however, not be reinstated in ONS’s labour market statistics First Release because of concerns over their appropriateness as a labour market indicator.

* From June 2005, data on Jobcentre Plus vacancies, including inflows, unfilled stocks and outflows, are being published for an enhanced range of geographies including super output areas, parliamentary constituencies, local authority districts, and wards. Limited back data are available, but the figures are not directly comparable with those previously published.

* Interpretation of the data needs to take account of significant changes to Jobcentre Plus procedures for dealing with employers’ vacancies.

Introduction

Jobcentre vacancy statistics have traditionally been produced as a by-product of the administrative systems used for handling vacancies notified by employers. The figures have been published, for example, in the tables section of Labour Market Trends (see Tables G. 11, G.12 and G.13).

In September 2001 ONS, with the agreement of the Department for Work and Pensions (DWP), deferred the publication of Jobcentre Plus vacancy statistics because of distortions in the data from May 2001 onwards. In September 2002 publication of notified vacancies (inflows) only was resumed on Nomis®.

The purposes of this article follow below.

* To publicise the restoration on Nomis® of Jobcentre Plus vacancy stock and outflow data together with improvements to the range of available geographies and explain how these have been derived.

* To announce the publication of National Statistics on Jobcentre Plus vacancies in the quarterly DWP Statistical Summary in light of the decision by the National Statistician not to restore the data to ONS’s labour market statistics First Release because of concerns over their suitability as a labour market indicator.

* To summarise technological and operational changes to vacancy taking and handling within Jobcentre Plus, including the advent of e-channels and what this means for the interpretation of the vacancy series.

* To provide summary time series analyses and illustrate the dynamics between vacancy inflows, stocks and outflows.

* To highlight the uses and limitations of the datasets.

* To summarise the position in regard to the derivation and dissemination of Northern Ireland Jobcentre vacancy statistics.

Changes to Jobcentre Plus vacancy taking and handling

Interpretation of the Jobcentre vacancy time series requires an understanding of the changes that have taken place within Jobcentre Plus both for vacancy taking and vacancy handling and the impact these have had on flow and stock data.

The past few years has seen a significant period of change within Jobcentre Plus in both the taking and handling of employers’ vacancies. A previous Labour Market Trends article explained how changes to vacancy taking resulting from the introduction of Employer Direct in 2001 (which moved vacancy taking and follow up away from local Jobcentres to a virtual network of contact centres), affected the vacancy series (see pp363-68, July 2003). Since then, other elements of Jobcentre Plus modernisation of services to employers have been introduced. The strategy has been to offer employers a wider choice of channels through which to notify their vacancies. In addition to telephone, e-mail and fax options offered by Employer Direct, use of e-channels has been developed through Employer Direct online (allowing employers to notify and update vacancies directly to Jobcentre Plus via the internet). For employers and agencies who do not require the additional services offered by Jobcentre Plus of matching clients to vacancies, a Job Warehouse has been developed.

As the name suggests, this allows selected recruiters to provide information on vacancies to a central database that can be accessed by jobseekers. Vacancies notified by this route are not recorded as Jobcentre Plus vacancies and hence are not included in the statistical series. The take up of both Employer Direct online and Job Warehouse is small but growing, with the national implementation of Employer Direct online in March 2005. The introduction of Job Warehouse is slower and dependent on ecapabilities within recruiting organisations and their suitability for this channel. Both initiatives illustrate how vacancy inflow volumes could be influenced by operational change either through boosting the market share of all vacancies handled by Jobcentre Plus by attracting new business, or by displacing inflows through the availability of Job Warehouse.

These technological changes have been accompanied by operational change within Jobcentre Plus to establish closer customer relationships with employers. From April 2003 Jobcentre Plus undertook a more significant focus on marketing of Jobcentre Plus services to employers (through telemarketing campaigns to attract more business, establishment of a national sales force and tiers of account managers, and introduction of targets for the number of notified vacancies). In addition, a number of Service Level Agreements have been established with agencies. The impact of this on the vacancy series has been an increase in notified vacancies in certain sectors of the economy, for example, public administration, health and social work, and real estate and business activities including agency business.

The accuracy of both the stock and outflow series is dependent on timely and effective follow up. Follow up (usually in the form of a telephone call to the employer) is the means by which a vacancy is kept open (unfilled), suspended or closed. Delays to follow up or ineffective procedures result in a build-up of unfilled vacancies and, at a time when inflow levels have been strong, the pressure for timely and efficient follow up builds. An improved process of vacancy follow up with employers has been adopted throughout Jobcentre Plus from June 2004 through the use of Intelligent Follow Up. This offers a more flexible system of follow up by agreeing with the employer at the time the vacancy is taken when the follow up will be conducted. It involves recontacting the employer more rigorously if they are unavailable, and ultimately closing vacancies down in cases where follow up has not been successful in order to avoid a build-up of vacancy deadwood in the operational systems.

It is possible that further change to Jobcentre Plus vacancy taking and follow up will have significant impact on future vacancy statistics and the comparability of the series over time.

Enhancements to Jobcentre Plus vacancy statistics published on Nomis®

From June 2005 a much wider set of Jobcentre Plus vacancy data will be available on Nomis®. The key improvements are described below.

Improved range of geographies

Recent availability of postcode level data has meant that a range of previously unavailable geographies will be an option for users. Data by parliamentary constituencies, super output areas, local authority districts and wards will be available for inflow, stock, and outflow datasets. In the main (95 per cent of cases), the postcode of the vacancy will be used to generate super output area, local authority district and parliamentary constituency. Where this is missing or misrecorded and in the case of speculative placings3 the employers’ postcode will be used (4 per cent of cases). In the absence of both the vacancy and employer postcodes (1 per cent of cases), the postcode of the Jobcentre Plus office designated as owning the vacancy is used to ensure that all vacancies are allocated a valid postcode.

Restoration of stock of unfilled vacancies data

Previously published stock data implicitly defined unfilled vacancies as comprising both ‘live’ and ‘suspended’ vacancies. Reinstated stock data from June 2005 will differentiate between live vacancies (those available to jobseekers) and suspended vacancies (those no longer available to jobseekers but not yet closed). Suspended vacancies comprise about one third of all Jobcentre Plus unfilled vacancies recorded in the Jobcentre Plus system.

In addition, data by duration and a backseries of monthly data will be available on Nomis®. Users will – for the first time – have the option of defining vacancy stock as inclusive or exclusive of suspended vacancies and for all durations or by specific duration bands; for example, it may be appropriate just to consider live vacancies that have been unfilled for six months or less. Generally, users are recommended to define stocks as only the live element of unfilled vacancies since it is only this element that are available to jobseekers. For Nomis® queries, the system will default to analyses of live rather than total unfilled vacancies.

Restoration of vacancy outflow data

To complete the picture, data on monthly vacancy outflows will also be reinstated on Nomis®. Within all outflows, data on filled and withdrawn vacancies will be available. Users of the statistics should note that a vacancy is only shown as filled if it is as a result of Jobcentre Plus submitting a client to that vacancy and the client subsequently gaining the job. Many vacancies notified to Jobcentre Plus will also have been advertised by employers through other recruitment channels (such as local newspapers and private recruitment agencies) and may have been filled through these routes. Subject to lags in recording, these vacancies will be shown as other outflow rather than filled vacancies in the published Nomis® statistics. This is consistent with the way in which outflows were defined prior to their suspension on Nomis®.

DWP release of National Statistics

From the 15 June 2005 quarterly release of the DWP Statistical Summary, a set of new National Statistics of Jobcentre Plus vacancies will be available online at www. dwp.gov. uk/asd/statistics. Coverage will be for Great Britain and the data will be sourced from the Nomis® published figures. The tables will give a measure of vacancy volumes for monitoring the performance of Jobcentre Plus, with figures shown over time, by Jobcentre Plus region, and by occupation.

Jobcentre vacancy data for Northern Ireland are not classified as National Statistics nor are they currently available through Nomis®. However, data are published in the Northern Ireland Labour Market Report available on the Department of Enterprise, Trade and Investment’s website.

Trends in vacancy statistics

To give a perspective on recent trends in the vacancy series and the impact of changes in Jobcentre Plus vacancy handling, Figures 1 to 4 track trends in the time series since January 2000.

Vacancy inflows

Average monthly vacancies notified have risen year on year from 225,000 per month in 2000, to 264,000 per month in 2002, to 301,000 per month in 2004. Figure 1 shows seasonally adjusted monthly notified vacancies overlaid with a series of timelines indicating operational change within Jobcentre Plus as it relates to vacancy taking. Periods of significant operational change (the introduction of Employer Direct in 2001 and the focus on target setting and marketing in 2003) coincide well with step changes in the monthly inflow volumes of notified vacancies. The impact of Employer Direct is described in much more detail in a previous Labour Market Trends article, which concluded that recorded vacancy inflows increased by about 20 per cent as a result (see pp363-68, July 2003).

Stock of unfilled vacancies

During the period January 2001 to December 2004 the Jobcentre Plus stock of unfilled vacancies rose from 340,000 to just over one million. Discounting suspended vacancies brings the stock figure down to 750,000, which remains high in comparison to ONS survey wholeeconomy estimates. Clearly, Jobcentre Plus does not have a 100 per cent market share of the vacancy market (2002 estimates derived from the Vacancy Survey suggested that, at this time, the market share lay between a third and a half, see pp349-361, Labour Market Trends, July 2003). Users of both statistical series should note that the two sets of figures are not directly comparable. While the ONS estimates are derived from a survey of employers’ unfilled vacancies at a point in time, the Jobcentre Plus figures will – since they are derived from administrative data – always reflect the fact that until a vacancy is followed up and either closed (filled, cancelled, or withdrawn) or suspended it continues to be recorded as unfilled. Since follow up is not instantaneous, the Jobcentre Plus stock series will always reflect natural lags in the system. The main use of the stock data is for crosssectional analyses as an indicator of the range of jobs available. Comparability of the series over time has been compromised by the changes to vacancy handling and follow up procedures.

Vacancy outflows

To complete the picture and shed light on the increases in vacancy stocks, Figure 3 shows the trend in median duration of vacancies at the time of outflow for all outflows and separately for filled vacancies and non-filled vacancies.

For those vacancies filled by Jobcentre Plus, median duration (measured from the date of vacancy notification to the date of vacancy closure) rose from an average of two weeks in 2000 to three weeks in 2004. This shift partly reflects definitional change introduced in 2001 whereby a vacancy could only be recorded as filled when a jobseeker had started employment rather than simply having received an offer of a job. The median duration of vacancies which are closed but not filled by Jobcentre Plus increased from six weeks in 2000 to 11 weeks in 2004. This, coupled with the fact that about three-quarters of vacancy outflows are such non-filled vacancies, is the major determinant of the increases in vacancy stocks.

Uses and limitations of Jobcentre Plus vacancy statistics

Based on Jobcentre Plus’s labour market computer system, the 100 per cent administrative vacancy data has traditionally served a number of purposes.

* At the macro level, the time series of seasonally adjusted notified vacancies has been used as a labour market indicator.

* It has served as a spatial indicator of local labour market demand, particularly for detailed analysis of the stock of unfilled vacancies in specific occupations, industrial sectors and local areas.

* Within DWP it has been used as a measure of Jobcentre Plus performance both in terms of vacancy taking (inflows) and vacancy management (stocks and outflows).

The jobcentre administrative vacancy data have a number of advantages over survey data:

* they give 100 per cent coverage (for Jobcentre Plus vacancies), allowing for robust local level analysis including by four-digit Standard Occupational Classification codes and twodigit Standard Industrial Classification codes;

* they have no sampling errors;

* and they are cheap and relatively easy to produce.

But they also have some drawbacks.

* They are not whole-economy figures. Coverage is just for those vacancies notified to Jobcentre Plus and as such represent a market share of vacancies throughout the whole economy. This proportion varies over time, according to the occupation and industry of the vacancies, and according to geographical location.

* As Figure 1 shows, the time series is susceptible to discontinuities arising from changes to vacancy handling within Jobcentre Plus (best demonstrated with the move to a virtual network of contact centres when Employer Direct rolled out), marketing strategies/target measures and definitional change, making comparability over time difficult.

* They are affected, in the case of vacancy stocks, by delayed live updating of the administrative database pending follow up.

Series dynamics

In most cases the relationship between the three series is that the stock at the end of a period is equal to the stock at the beginning of the period plus the inflows and minus the outflows during that period. However, this relationship only holds when the stock is defined as both the live and suspended elements and no longer holds when data is standardised or seasonally adjusted.

The stock of unfilled vacancies at a point in time is composed of many previous months’ vacancy inflows which have not – as yet – been translated into vacancy outflows. Figure 4 shows the typical progression for one month’s cohort of vacancy inflows (September 2003) tracked over time. From the point of notification, the inflow cohort either has the status of unfilled stock, filled outflow, or non-filled outflow.

Northern Ireland statistics

Publication of Jobcentre vacancy data for Northern Ireland resumed in March 2005 through the Department of Enterprise, Trade and Investment’s website. Data had previously been suspended because of computing anomalies that resulted in dubious outputs. These have now been identified and corrected. Vacancy statistics are extracted from the Department for Employment and Eearning’s Client Management System each month. Headline data on vacancy inflows, stocks, and outflows are published. Data are not classified as National Statistics, nor are they currently available via Nomis®. In broad terms, Northern Ireland Jobcentre vacancies comprise around 1 per cent to 2 per cent of the UK total.

Further information

For further information, contact:

Russ Bentley, Information Directorate, Department for Work and Pensions, Level 2, Kings Court, 80 Hanover Way, Sheffield S3 7UF, E-mail: russ.bentley@dwp.gov.uk, Tel: 0114 209 8236.

Notes

1 For similar reasons Tables G.11, G.12 and G.13 in the tables section of Labour Market Trends, which show previously published jobcentre vacancy data up to April 2001, will be discontinued from July 2005.

2 Suspended vacancies are those neither closed nor currently available to jobseekers. In the majority of cases vacancies are suspended because the submissions limit or the closing date has been reached but there is outstanding follow up of the vacancy required before it can be closed.

3 Speculative placings are those for which a vacancy has not been notified to Jobcentre Plus in the normal sense, but in which Jobcentre Plus speculatively refer a client to an employer in any case, and the client subsequently gains a job (placing).

By Russ Bentley, Information Directorate, Department for Work and Pensions

Copyright The Controller of Her Majesty’s Stationery Office Jun 2005

Provided by ProQuest Information and Learning Company. All rights Reserved