UK labour market: November 2018

• Estimates from the Labour Force Survey show that, between April to June 2018 and July to September 2018, the number of people in work and the number of unemployed people both increased but the number of people aged from 16 to 64 years not working and not seeking nor available to work (economically inactive) was little changed.

• There were 32.41 million people in work, 23,000 more compared with April to June 2018 and 350,000 more than for a year earlier.

• The employment rate (the proportion of people aged from 16 to 64 years who were in work) was 75.5%, little changed compared with April to June 2018 but higher than for a year earlier (75.0%).

• There were 1.38 million unemployed people (people not in work but seeking and available to work), 21,000 more than for April to June 2018 but 43,000 fewer than for a year earlier.

• The unemployment rate (the number of unemployed people as a proportion of all employed and unemployed people) was 4.1%, slightly higher than for April to June 2018 but lower than for a year earlier (4.3%).

• There were 8.74 million people aged from 16 to 64 years who were economically inactive (not working and not seeking nor available to work), little changed compared with April to June 2018 but 147,000 fewer than for a year earlier.

• The economic inactivity rate (the proportion of people aged from 16 to 64 years who were economically inactive) was 21.2%, unchanged compared with April to June 2018 but lower than for a year earlier (21.6%).

• Latest estimates show that average weekly earnings for employees in Great Britain in nominal terms (that is, not adjusted for price inflation) increased by 3.2% excluding bonuses, and by 3.0% including bonuses, compared with a year earlier.

• Latest estimates show that average weekly earnings for employees in Great Britain in real terms (that is, adjusted for price inflation) increased by 0.9% excluding bonuses, and by 0.8% including bonuses, compared with a year earlier.

Table 1, Figure 1a and Figure 1b show the latest estimates, for July to September 2018, for employment, unemployment and economic inactivity and show how these estimates compare with the previous quarter (April to June 2018) and the previous year (July to September 2017). Comparing the estimates for July to September 2018 with those for April to June 2018 provides the most robust short-term comparison. See Section 3 of this statistical bulletin for more information.

About labour market statuses

Everybody aged 16 years or over is either employed, unemployed or economically inactive. The employment estimates include all people in paid work including those working part-time. People not working are classed as unemployed if they have been looking for work within the last four weeks and are able to start work within the next two weeks. A common misconception is that the unemployment statistics are a count of people on benefits; this is not the case as they include unemployed people not claiming benefits.

Jobless people who have not been looking for work within the last four weeks or who are unable to start work within the next two weeks are classed as economically inactive. Examples of economically inactive people include people not looking for work because they are students, looking after the family or home, because of illness or disability or because they have retired.

Making comparisons with earlier data derived from the Labour Force Survey

Estimates of employment, unemployment, economic inactivity, hours worked and redundancies are derived from the Labour Force Survey (LFS), a survey of households. The most robust estimates of short-term movements in these estimates are obtained by comparing the estimates for July to September 2018 with the estimates for April to June 2018, which were first published on 14 August 2018. This provides a more robust estimate than comparing with the estimates for June to August 2018. This is because the July and August 2018 data are included within both estimates, so effectively observed differences are those between the individual months of June and September 2018. The LFS is sampled such that it is representative of the UK population over a three-month period, not for single month periods.

Accuracy and reliability of survey estimates

Most of the figures in this statistical bulletin come from surveys of households or businesses and are therefore estimates rather than precise figures. Further information is available in the Quality and methodology section of this statistical bulletin.

Where to find explanatory information

A Guide to labour market statistics, which includes a Glossary, is available.

Things you need to know about employment

Employment measures the number of people in paid work and differs from the number of jobs because some people have more than one job.

Employment mainly consists of employees and self-employed people but there are also two minor categories – unpaid family workers and people on government-supported training and employment programmes.

Unpaid family workers are people who work in a family business who do not receive a formal wage or salary but benefit from the profits of that business.

The government-supported training and employment programmes series does not include all people on these programmes; it only includes people engaging in any form of work, work experience or work-related training who are not included in the employees or self-employed series. People on these programmes not engaging in any form of work, work experience or work-related training are not included in the employment estimates; they are classified as unemployed or economically inactive.

Commentary

The proportion of people aged from 16 to 64 years in work is known as the employment rate. Figure 2 shows the employment rates for all people, and for men and women, aged from 16 to 64 years since comparable records began in 1971. The lowest employment rate for all people aged from 16 to 64 years was 65.6% in 1983, during the economic downturn of the early 1980s. The employment rates for both men and women have been generally increasing since early 2012. For the latest time period, July to September 2018, the employment rate for all people aged from 16 to 64 years was 75.5%, up from 75.0% for a year earlier, but little changed compared with April to June 2018.

Figure 3 looks in more detail at the employment rate for all people aged from 16 to 64 years for the last five years.

For July to September 2018:

• 75.5% of all people aged from 16 to 64 years were in work, up from 75.0% for a year earlier, but little changed compared with April to June 2018

• 80.0% of men aged from 16 to 64 years were in work; up from 79.5% for a year earlier but slightly lower than for April to June 2018 (80.1%)

• 71.0% of women aged from 16 to 64 years were in work, up from 70.6% for a year earlier but unchanged compared with April to June 2018

The increase in the employment rate for women over the last few years has been partly due to ongoing changes to the State Pension age for women, resulting in fewer women retiring between the ages of 60 and 65 years.

For July to September 2018, there were 32.41 million people aged 16 years and over in work, 23,000 more compared with April to June 2018 and 350,000 more than for a year earlier.

Figure 4 shows that the annual increase in the number of people in work (350,000) was entirely due to more people in full-time employment (416,000).

Comparing the estimates for employees and self-employed people for July to September 2018 with those for a year earlier:

• employees increased by 425,000 to 27.50 million (84.9% of all people in work)

• self-employed people decreased by 59,000 to 4.75 million (14.7% of all people in work)

Employees and self-employed people do not account for all people in employment as there are two minor additional categories (unpaid family workers and people on government-supported training and employment programmes). Further information is available at the beginning of this section.

Comparing the estimates for employment by nationality (not seasonally adjusted) for July to September 2018 with those for a year earlier:

• UK nationals working in the UK increased by 448,000 to 29.00 million

• EU nationals working in the UK fell by 132,000 to 2.25 million (the largest annual fall since comparable records began in 1997)

• Non-EU nationals working in the UK increased by 34,000 to 1.24 million

Detailed information on employment by nationality and by country of birth is available at UK and non-UK people in the labour market.

Where to find data about employment

Employment estimates are available at Tables 1 and 3 of the PDF version of this statistical bulletin and at datasets A02 SA and EMP01 SA.

Estimates of employment, unemployment and economic inactivity for UK and non-UK workers are available at Non-UK workers in the labour market and at datasets A12 and EMP06.

International comparisons of employment rates are available at Table 17 of the PDF version of this statistical bulletin and at dataset A10.

Historic estimates of employment (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheets A49 and A50).

Things you need to know about public and private sector employment

Public sector employment measures the number of people in paid work in the public sector. The public sector comprises central government, local government and public corporations. Estimates of public sector employment are obtained from information provided by public sector organisations.

Private sector employment is estimated as the difference between total employment, sourced from the Labour Force Survey, and public sector employment.

Comparisons of public and private sector employment over time are impacted by changes to the composition of these sectors. For example, if a publicly owned body is privatised, public sector employment will fall and private sector employment will increase by an equivalent amount. This is known as a reclassification effect. At Table 4 of the PDF version of this statistical bulletin and at dataset EMP02 we therefore publish estimates of public and private sector employment excluding the effects of major, but not all, reclassifications alongside estimates of total public and private sector employment.

Commentary

For June 2018, there were 5.34 million people employed in the public sector, 4,000 fewer than for March 2018; this small fall was entirely due to the transfer of Welsh housing associations to the private sector. Excluding the Welsh housing associations transfer, the number of people employed in the public sector increased by 16,000 between March and June 2018.

Between June 2017 and June 2018, public sector employment fell by 122,000; this large fall was entirely due to the transfer of housing associations in England (in December 2017) and in Wales (in June 2018) to the private sector. Excluding the housing associations transfer, the number of people employed in the public sector increased by 34,000 between June 2017 and June 2018, mainly due to more people working for the National Health Service.

For June 2018, there were 1.65 million people employed in the National Health Service (30.9% of all people employed in the public sector) and there were 1.51 million people employed in public sector education (28.3% of all people employed in the public sector). As shown in Figure 5, over the last five years, employment in the National Health Service has steadily increased while employment in public sector education has remained broadly flat.

Where to find data about public and private sector employment

Public and private sector employment estimates are available at Tables 4 and 4(1) of the PDF version of this statistical bulletin and at datasets EMP02 and EMP03.

Further information on public sector employment is available in the Public sector employment release.

Historic estimates of public sector employment (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheet A51).

Things you need to know about actual hours worked

Actual hours worked measures the number of hours worked in the economy. Changes in actual hours worked reflect changes in the number of people in employment and the average hours worked by those people.

Commentary

Between April to June 2018 and July to September 2018, total hours worked increased (by 10.7 million) to 1.04 billion. This reflected an increase of 23,000 in the number of people in employment (as explained in Section 4 of this statistical bulletin) and an increase in average weekly hours worked, particularly by those working full-time.

For July to September 2018:

• people worked, on average, 32.2 hours per week, 0.3 hours more than for April to June 2018 and 0.1 hours more than for a year earlier

• people working full-time worked, on average, 37.4 hours per week in their main job, 0.3 hours more than for April to June 2018 but unchanged compared with a year earlier

• people working part-time worked, on average, 16.3 hours per week in their main job, 0.1 hours more than for April to June 2018 but unchanged compared with a year earlier

Where to find data about hours worked

Hours worked estimates are available at Tables 7 and 7(1) of the PDF version of this statistical bulletin and at datasets HOUR01 SA and HOUR02 SA.

Historic estimates of hours worked (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheet A54).

Things you need to know about workforce jobs

Workforce jobs measures the number of filled jobs in the economy. The estimates are mainly sourced from employer surveys. Workforce jobs is a different concept from employment, which is sourced from the Labour Force Survey, as employment is an estimate of people in work and some people have more than one job.

Commentary

For June 2018, there were 35.20 million workforce jobs, 83,000 more than for March 2018 and 132,000 more than for a year earlier.

As shown in Figure 7, while most industrial sectors showed increases in the number of jobs between June 2017 and June 2018, the number of jobs in wholesaling, retailing and motor vehicle repairs fell by 73,000. This was the largest annual fall in the number of jobs in that sector since June 2010.

Looking at a longer-term comparison, between June 1978 (when comparable records began) and June 2018:

• the proportion of jobs accounted for by the manufacturing, and mining and quarrying sectors fell from 26.4% to 7.9%

• the proportion of jobs accounted for by the services sector increased from 63.2% to 83.3%

Where to find data about workforce jobs

Jobs estimates are available at Tables 5 and 6 of the PDF version of this statistical bulletin and at datasets JOBS01 and JOBS02.

While comparable estimates for workforce jobs by industry begin in 1978, some historical information back to 1841, not comparable with the latest estimates, are available from 2011 Census Analysis, 170 years of industry.

Historic estimates of jobs by industry (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheet A53).

Things you need to know about average weekly earnings

Average weekly earnings measures money paid per week, per job to employees in Great Britain in return for work done, before tax and other deductions from pay. The estimates do not include unearned income, benefits in kind or arrears of pay. The estimates do not include earnings of self-employed people.

The estimates are not just a measure of pay rises as they also reflect changes in the number of paid hours worked and the impact of employees paid at different rates joining and leaving individual businesses. The estimates also reflect changes in the overall structure of the workforce; for example, more low-paid jobs in the economy would have a downward effect on the earnings growth rate.

Estimates are available for both total pay (which includes bonuses) and for regular pay (which excludes bonus payments). Estimates are available in both nominal terms (not adjusted for consumer price inflation) and real terms (adjusted for consumer price inflation). The estimates of average weekly earnings in real terms are calculated by deflating the nominal earnings estimates by the Consumer Prices Index including owner occupiers’ housing costs (CPIH), our preferred measure of consumer price inflation. The CPIH figures and, consequently, the estimates of average weekly earnings in real terms, start in January 2005. Comparable estimates of average weekly earnings in nominal terms start in January 2000.

Commentary

For September 2018, average regular pay (excluding bonuses), before tax and other deductions from pay, for employees in Great Britain was:

• £493 per week in nominal terms (that is, not adjusted for price inflation), up from £478 per week for a year earlier

• £463 per week in constant 2015 prices (that is, adjusted for price inflation), up from £458 per week for a year earlier, but £10 lower than the pre-downturn peak of £473 per week for March 2008

Looking at annual growth rates for regular pay (excluding bonuses), between July to September 2017 and July to September 2018:

• regular pay in nominal terms increased by 3.2%; the annual growth rate has not been higher since October to December 2008

• regular pay in real terms increased by 0.9%, the annual growth rate has not been higher since October to December 2016

For September 2018, average total pay (including bonuses), before tax and other deductions from pay, for employees in Great Britain was:

• £524 per week in nominal terms, up from £510 per week for a year earlier

• £493 per week in constant 2015 prices, up from £490 per week for a year earlier, but £29 lower than the pre-downturn peak of £522 per week for February 2008

Looking at annual growth rates for total pay (including bonuses), between July to September 2017 and July to September 2018:

• total pay in nominal terms increased by 3.0%, the annual growth rate has not been higher since July to September 2015

• total pay in real terms increased by 0.8%, the annual growth rate has not been higher since October to December 2016

Where to find data about average weekly earnings

Estimates of average weekly earnings in nominal terms (that is, not adjusted for consumer price inflation) are available at Tables 13, 14 and 15 of the PDF version of this statistical bulletin and at datasets EARN01, EARN02 and EARN03.

Estimates of average weekly earnings in real terms (that is, adjusted for consumer price inflation) are available at Table 16 of the PDF version of this statistical bulletin and at dataset EARN01.

While comparable records for average weekly earnings start in 2000, modelled estimates of average weekly earnings in nominal terms back to 1963 (which do not have National Statistics status) are available at dataset EARN02.

Historic estimates (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheets A47 and A48).

Where to find more information about earnings

The Annual Survey of Hours and Earnings (ASHE), published on 25 October 2018, provides more detailed data for 2018.

Experimental earnings statistics based on Pay As You Earn (PAYE) administrative data have been published by HM Revenue and Customs.

Things you need to know about labour disputes

Labour disputes estimates measures strikes connected with terms and conditions of employment.

Commentary

For September 2018:

• there were 9,000 working days lost from 11 stoppages

• 3,000 people took strike action

For the 12 months ending September 2018:

• there were 265,000 working days lost from 74 stoppages and 35,000 people took strike action

• there were 245,000 working days lost in the private sector

• there were 20,000 working days lost in the public sector

Since monthly records began in 1931:

• the highest cumulative 12-month estimate for working days lost was 32.2 million for the 12 months to April 1980

• the lowest cumulative 12-month estimate for working days lost was 143,000 for the 12 months to March 2011

Where to find data about labour disputes

Labour disputes estimates are available at Table 18 of the PDF version of this statistical bulletin and at dataset LABD01.

Where to find more information about labour disputes

The labour disputes annual article provides more detailed information. The most recent edition of this article was published on 30 May 2018.

Things you need to know about unemployment

Unemployment measures people without a job who have been actively seeking work within the last four weeks and are available to start work within the next two weeks.

The unemployment rate is not the proportion of the total population who are unemployed. It is the proportion of the economically active population (those in work plus those seeking and available to work) who are unemployed. This follows guidelines specified by the International Labour Organisation and it ensures that UK unemployment statistics are broadly comparable with those published by other countries.

Commentary

The proportion of the economically active population (all employed and unemployed people) who are unemployed is known as the unemployment rate. As shown in Figure 11 (which shows unemployment rates for all people and for men and women), the lowest unemployment rate for all people since comparable records began in 1971 was 3.4% in late 1973 to early 1974 and the highest rate, 11.9%, was in 1984 during the economic downturn of the early 1980s. The unemployment rate for all people for the latest time period, July to September 2018, was 4.1%.

Figure 12 looks in more detail at unemployment rates for the last five years.

For July to September 2018:

• the unemployment rate for all people was 4.1%

• the unemployment rate for men was 4.2%

• the unemployment rate for women was 4.0%; the joint-lowest since comparable records began in 1971

For July to September 2018, there were:

• 1.38 million unemployed people, 21,000 more than for April to June 2018 but 43,000 fewer than for a year earlier

• 750,000 unemployed men, 29,000 more than for April to June 2018 but 20,000 fewer than for a year earlier

• 631,000 unemployed women, 8,000 fewer than for April to June 2018 and 23,000 fewer than for a year earlier

Looking at unemployment by how long people have been out of work and seeking work, for July to September 2018, there were:

• 798,000 people who had been unemployed for up to six months, 21,000 fewer than for a year earlier

• 214,000 people who had been unemployed for between six months and one year, little changed compared with a year earlier

• 369,000 people who had been unemployed for over one year, 21,000 fewer than for a year earlier

Where to find data about unemployment

Unemployment estimates for the UK are available at Table 9 of the PDF version of this statistical bulletin and at dataset UNEM01 SA.

Estimates of employment, unemployment and economic inactivity for UK and non-UK workers are available at Non-UK workers in the labour market and at datasets A12 and EMP06.

Historic estimates of unemployment (which do not have National Statistics status) have been published by the Bank of England in the spreadsheet A millennium of macroeconomic data version 3.1 (at worksheets A49 and A50).

International comparisons of unemployment rates are available at Table 17 of the PDF version of this statistical bulletin and at dataset A10.

Things you need to know about economic inactivity

Economically inactive people are not in employment but do not meet the internationally accepted definition of unemployment because they have not been seeking work within the last four weeks and/or they are unable to start work within the next two weeks.

Commentary

The proportion of people, aged from 16 to 64 years, not in work and neither seeking nor available to work is known as the economic inactivity rate. Figure 13 shows that, since comparable records began in 1971, the economic inactivity rate for all people aged from 16 to 64 years has been generally falling (although it increased during economic downturns) due to a gradual fall in the economic inactivity rate for women. The economic inactivity rate for men has been gradually rising.

Figure 14 looks in more detail at the economic inactivity rate for all people aged from 16 to 64 years since comparable records began in 1971. It increased during the economic downturn of the early 1980s, reaching a record high of 25.9% in 1983. As the economy improved in the late 1980s, it resumed its downward path, before the economic downturn of the early 1990s drove it back up again. Following an increase in the economic inactivity rate during the economic downturn of 2008 to 2009, it again resumed a generally downward path.

Looking at the latest economic inactivity rates, for people aged from 16 to 64 years, for July to September 2018:

• the rate for all people was 21.2%, unchanged compared with April to June 2018 but lower than for a year earlier (21.6%)

• the rate for men was 16.3%, slightly lower than for April to June 2018 (16.4%) and lower than for a year earlier (16.8%)

• the rate for women was 26.0%, virtually unchanged compared with April to June 2018 but lower than for a year earlier (26.3%)

For July to September 2018, there were 8.74 million people aged from 16 to 64 years not in work and neither seeking nor available to work (known as economically inactive). This was:

• little changed compared with April to June 2018

• 147,000 fewer than for a year earlier

Looking in more detail at the 8.74 million people aged from 16 to 64 years who were economically inactive for July to September 2018, the three largest categories were students (27.3% of the total), people looking after the family or home (23.4% of the total) and long-term sick (22.8% of the total):

• there were 2.38 million people who were not looking for work because they were studying, 17,000 fewer than for a year earlier

• there were 2.04 million people who were not looking for work because they were looking after the family or home, 23,000 fewer than for a year earlier

• there were 1.99 million people who were not looking for work due to long-term sickness, 53,000 fewer than for a year earlier

The fourth-largest category within economic inactivity for those aged from 16 to 64 years was retired (12.8% of the total). There were 1.12 million people who were not looking for work because they had retired, 66,000 fewer than for a year earlier.

Figure 15 shows that the categories showing the largest falls over the five-year period from July to September 2013 to July to September 2018 were looking after the family and home (down 282,000) and retired (down 233,000). This reflects ongoing changes to the State Pension age for women, resulting in fewer women retiring between the ages of 60 and 65 years, as well as more women in younger age groups participating in the labour market.

Where to find data about economic inactivity

Economic inactivity estimates are available at Tables 1 and 11 of the PDF version of this statistical bulletin and at datasets A02 SA and INAC01 SA.

Estimates of employment, unemployment and economic inactivity for UK and non-UK workers are available at Non-UK workers in the labour market and at datasets A12 and EMP06.

Things you need to know about young people in the labour market

This section looks at people aged from 16 to 24 years. It is a common misconception that all people in full-time education are classified as economically inactive. This is not the case as people in full-time education are included in the employment estimates if they have a part-time job and are included in the unemployment estimates if they are seeking part-time work.

Commentary

Figure 16 shows the number of people aged from 16 to 24 years who were employed, unemployed and economically inactive for July to September 2018, broken down by educational status.

Figure 17 shows that the UK household population is falling for those aged from 16 to 24 years. It also shows that, for people aged from 16 to 24 years, between July to September 2017 and July to September 2018:

• the number of people in employment fell by 15,000 to 3.81 million

• the number of unemployed people fell by 41,000 to 476,000

• the number of economically inactive people fell by 36,000 to 2.71 million (most of whom were full-time students)

For July to September 2018, the unemployment rate for those aged from 16 to 24 years was 11.1%, lower than for a year earlier (11.9%). The unemployment rate for those aged from 16 to 24 years has been consistently higher than that for older age groups since comparable records began in 1992.

Between March to May 1992 and July to September 2018, the proportion of people aged from 16 to 24 years who were in full-time education increased substantially from 26.2% to 44.5%. This has impacted on the youth unemployment rate because the increase in the number of young people going into full-time education has reduced the size of the economically active population (those in work plus those seeking and available to work). A fall in the size of the economically active population leads to a higher unemployment rate (because the unemployment rate is the proportion of the economically active population who are unemployed, not the proportion of the total population who are unemployed).

Where to find data about young people in the labour market

Estimates for young people in the labour market are available at Table 12 of the PDF version of this statistical bulletin and at dataset A06 SA.

Where to find more information about young people in the labour market

Estimates for young people who were not in education, employment or training (NEET) for April to June 2018 were published on 23 August 2018. Estimates for July to September 2018 will be published on 22 November 2018.

Things you need to know about redundancies

The redundancies estimates measure the number of people who were made redundant or who took voluntary redundancy in the three months before the Labour Force Survey interviews.

Commentary

For July to September 2018, 84,000 people had become redundant in the three months before the Labour Force Survey interviews. This was:

• 14,000 fewer than for April to June 2018

• 18,000 fewer than for July to September 2017

• the lowest since comparable records began in March to May 1995

Where to find data about redundancies

Redundancies estimates are available at Table 22 of the PDF version of this statistical bulletin and at dataset RED01 SA.

Things you need to know about vacancies

Vacancies are defined as positions for which employers are actively seeking to recruit outside their business or organisation.

Commentary

There were 845,000 job vacancies for August to October 2018. This was:

• 14,000 more than for May to July 2018

• 44,000 more than for a year earlier

• the highest since comparable records began in 2001

There were 2.8 job vacancies per 100 employee jobs for August to October 2018. The industrial sector showing the largest vacancy rate was accommodation and food service activities (4.1 vacancies per 100 filled employee jobs). The sectors showing the smallest vacancy rates were construction and education (both 1.8 vacancies per 100 filled employee jobs).

Where to find data about vacancies

Vacancies estimates are available at Tables 19, 20 and 21 of the PDF version of this statistical bulletin and at datasets VACS01, VACS02 and VACS03.

Publication dates up to the end of 2019 are:

11 December 2018
22 January 2019
19 February 2019
19 March 2019
16 April 2019
14 May 2019
11 June 2019
16 July 2019
13 August 2019
10 September 2019
15 October 2019
12 November 2019
17 December 2019

The Labour market economic commentary is published every month on the same day as this release.

The Regional labour market statistics release is published every month on the same day as this release. Regional and local area labour market statistics are also available on our NOMIS website.

The Public sector employment release is published four times a year (in March, June, September and December) on the same day as this release.

The Young people not in education, employment or training (NEET) release is published four times a year (in February, May, August and November).

The Working and workless households in the UK release is published four times a year.

The Annual Survey of Hours and Earnings (ASHE) release is published annually. The most recent release was published on 25 October 2018.

The Business Register and Employment Survey release is published annually. The most recent release was published on 27 September 2018.

An article on Sickness absence in the labour market was published on 30 July 2018.

An article on Trends in self-employment in the UK was published on 7 February 2018.

Revisions

Estimates for the most recent time periods are subject to revision due to the receipt of late and corrected responses to business surveys and revisions to seasonal adjustment factors, which are re-estimated every month. Estimates are subject to longer-run revisions, on an annual basis, resulting from reviews of the seasonal adjustment process. Estimates derived from the Labour Force Survey (a survey of households) are usually only revised once a year. Revisions to estimates derived from other sources are usually minor and are commented on in the statistical bulletin if this is not the case. Further information is available in the labour market statistics revisions policy.

One indication of the reliability of the main indicators in this statistical bulletin can be obtained by monitoring the size of revisions. Datasets EMP05, UNEM04 and JOBS06 record the size and pattern of revisions over the last five years. These indicators only report summary measures for revisions. The revised data itself may be subject to sampling or other sources of error. Our standard presentation is to show five years' worth of revisions (60 observations for a monthly series, 20 for a quarterly series).

Accuracy of the statistics: estimating and reporting uncertainty

Most of the figures in this statistical bulletin come from surveys of households or businesses. Surveys gather information from a sample rather than from the whole population. The sample is designed to allow for this, and to be as accurate as possible given practical limitations such as time and cost constraints, but results from sample surveys are always estimates, not precise figures. This means that they are subject to some uncertainty. This can have an impact on how changes in the estimates should be interpreted, especially for short-term comparisons.

There is a trade-off between sample size and sampling variability. As the number of people available in the sample gets smaller, the variability of the estimates that we can make from that sample size gets larger. What this means in practice is that estimates for small groups (for example, unemployed people aged from 16 to 17 years), which are based on quite small subsets of the Labour Force Survey sample, are less reliable and tend to be more volatile than estimates for larger aggregated groups (for example, the total number of unemployed people).

We can illustrate the level of uncertainty (also called “sampling variability”) around a survey estimate by defining a range around the estimate (known as a “confidence interval”) within which we think the real value that the survey is trying to measure lies. Confidence intervals are typically defined so that we can say we are 95% confident the true value lies within the range – in which case we refer to a “95% confidence interval”.

The number of unemployed people for July to September 2018 was estimated at 1,381,000, with a stated 95% confidence interval of plus or minus 73,000. This means that we are 95% confident that the true number of unemployed people was between 1,308,000 and 1,454,000. Again, the best estimate from the survey was that the number of unemployed people was 1,381,000.

As well as calculating precision measures around the numbers and rates obtained from the survey, we can also calculate them for changes in the numbers and rates. For example, for July to September 2018, the estimated change in the number of unemployed people since April to June 2018 was an increase of 21,000, with a 95% confidence interval of plus or minus 76,000. This means that we are 95% confident the actual change in unemployment was somewhere between an increase of 97,000 and a fall of 55,000, with the best estimate being an increase of 21,000. As the confidence interval for the change in unemployment (plus 97,000 to minus 55,000) includes zero, the estimated increase in unemployment of 21,000 is said to be “not statistically significant”.

In general, changes in the numbers (and especially the rates) reported in this statistical bulletin between three- month periods are small, and are not usually greater than the level that is explainable by sampling variability. In practice, this means that small, short-term movements in reported rates should be treated as indicative, and considered alongside medium-and long-term patterns in the series and corresponding movements in administrative sources, where available, to give a fuller picture.

Where to find data about uncertainty and reliability

Dataset A11 shows sampling variabilities for estimates derived from the Labour Force Survey.

Dataset JOBS07 shows sampling variabilities for estimates of workforce jobs.

The sampling variability of the three-month average vacancies level is around plus or minus 1.5% of that level.

Sampling variability information for average weekly earnings growth rates are available from the “Sampling Variability” worksheets within datasets EARN01 and EARN03.

Seasonal adjustment and uncertainty

Like many economic indicators, the labour market is affected by factors that tend to occur at around the same time every year; for example, school leavers entering the labour market in July and whether Easter falls in March or April. In order to compare movements other than annual changes in labour market statistics, such as since the previous quarter or since the previous month, the data are seasonally adjusted to remove the effects of seasonal factors and the arrangement of the calendar. All estimates discussed in this statistical bulletin are seasonally adjusted except where otherwise stated. While seasonal adjustment is essential to allow for robust comparisons through time, it is not possible to estimate uncertainty measures for the seasonally adjusted series.

Quality and Methodology Information reports

The Quality and Methodology Information reports contain important information on:

• the strengths and limitations of the data and how it compares with related data

• users and uses of the data

• how the output was created

• the quality of the output including the accuracy of the data

Labour Force Survey Quality and Methodology Information

Labour Force Survey performance and quality monitoring reports

Vacancy Survey Quality and Methodology Information

Workforce jobs Quality and Methodology Information

Average weekly earnings (AWE) Quality and Methodology Information

Labour disputes Quality and Methodology Information