Topic guide to: Redundancies

Labour Force Survey Redundancy Tables

Department: Office for National Statistics
Updated quarterly tables on redundancy, not seasonally adjusted. The tables cover redundancy for: total numbers, rates and re-employment rates; industry numbers and rates; regional numbers and rates; numbers and rates by age.

Search for:

• Current and past editions
• Future Editions
• All Editions

Northern Ireland Labour Market Report

Department: Enterprise, Trade and Investment (Northern Ireland)
Labour Market Statistics for Northern Ireland (employment, unemployment, earnings).

Search for:

• Current and past editions
• Future Editions
• All Editions

Back to the top

Redundancies are measured by the Labour Force Survey (LFS). The estimates measure the number of people, whether working or not working, who reported that they had been made redundant or had taken voluntary redundancy in the month of the reference week or in the two calendar months prior to this.

Redundancy levels and rates are published each month in the UK Labour Market Statistics First Release. The redundancy rate is based on the ratio of the redundancy level for the given three-month average time period to the number of employees in the previous quarter.

Estimates of redundancy levels and rates for the UK are available seasonally adjusted by sex, for rolling three-monthly average time periods.

More detailed estimates of redundancies (by region, industry, age and re-employment rates) are available not seasonally adjusted and for calendar quarters. Redundancies estimates are not available at local area level.

Estimates of redundancy levels and rates for the UK are available from 1995, but more detailed estimates are not available prior to 1997.

Back to the top

Labour Force Survey

The Labour Force Survey (LFS) is the largest regular household survey in the UK. The survey covers people resident in private households, National Health Service (NHS) accommodation and student halls of residence. It does not cover any other communal establishments. LFS interviews are conducted continuously throughout the year.

In any three-month period, a nationally representative sample of approximately 120,000 people aged 16 and over in around 53,000 households is interviewed. Each household is interviewed five times, at three monthly intervals.

The initial interview is conducted face-to-face by an interviewer visiting the address, except for residences north of the Caledonian Canal in Scotland (where face-to-face-interviews would be prohibitively expensive). The other interviews are conducted by telephone wherever possible. The survey asks a series of questions about respondents’ personal circumstances and their labour market activity.

The survey is conducted in five ‘waves’ so that in any one quarter, one wave will be receiving their first interview, one wave their second, and so on, with one wave receiving their fifth and final interview. Therefore, there is an 80 per cent overlap in the samples for each successive quarter.

Seasonal adjustment

Seasonal movements can occur in labour market data for a number of reasons including holidays and recruitment patterns. For example, a large number of people leave full-time education and enter the labour market in the summer.

To make it easier to identify the underlying movements in the labour market, changes due solely to seasonal influences are removed from many series. This process is known as seasonal adjustment. LFS estimates are seasonally adjusted using the X-12 ARIMA package developed by Statistics Canada. The seasonal adjustment of LFS data is usually reviewed annually.

Sampling Variability

Survey estimates are prone to sampling variability. The easiest way to explain this is by example. In the April to June 2008 period, there were estimated to be 7,876,000 economically inactive people of working age in the UK, according to the LFS (seasonally adjusted). These figures were published in August 2008.

If we drew another sample for the same period we could get a different result, which could be higher or lower. In theory, we could draw many samples, and each would give a different result. The spread of these results leads to sampling variability.

Once we know the sampling variability we can calculate a range of values about the sample estimate that represents the expected variation with a given level of assurance. This is called a confidence interval.

For a 95 per cent confidence interval we expect that, in 95 per cent of the samples, the confidence interval will contain the true value of inactivity that would have been obtained by surveying the entire population. For example, for April to June 2008, we can be 95 per cent confident that the true level of inactivity was within 123,000 of the estimate of 7,876,000 (that is, within the range 7,753,000 to 7,999,000).

Sampling variability also affects changes over time. Changes in inactivity between three-month periods are rarely greater than the level that is explainable by sampling variability. It is estimated that the number of inactive people of working age in the UK fell by 3,000 between January to March 2008 and April to June 2008 (seasonally adjusted). We can be 95 per cent confident that the true change lies in the range between a decrease of 90,000 and an increase of 84,000. It is more likely that inactivity decreased, rather than increased.

In general, the larger the number of people in the sample, the smaller the variation between estimates. Estimates based on the LFS for the whole of the UK are therefore more accurate than those for smaller geographical areas. Indications of sampling variability for national and regional LFS data are given in the Labour Market Statistics First Releases.

Back to the top

Back to the top