This article contains charts that compare the Labour Force Survey (LFS) single-month estimates with their equivalent three-month average rates for employment, unemployment and economic inactivity. The single-month estimates are derived from the same data source as the headline three-month figures but are not designated as National Statistics. Their use is restricted to helping to understand the movements in the headline three-month averages. For the three-month averages, the dates shown on the charts relate to the last month of the three (for example, March to May is indicated by May).
Model-based single-month estimates are now also produced and included within the data table accompanying this article. The model uses single-month wave-specific time series estimates for each variable, along with estimated variances, to produce modelled seasonally adjusted time series. These are currently experimental statistics.
This article also includes experimental estimates of labour market indicators broken down by individual weeks. While not providing robust estimates of labour market conditions, these can help users to understand the impact of the coronavirus (COVID-19) pandemic on a week-by week basis during the quarter.
LFS estimates presented in this article include interviews that took place during March to May 2020. Consequently, while some interviews relate to the period prior to the implementation of coronavirus social distancing measures, interviews from the final week of March 2020 onwards relate to the period following the government closure of schools, introduction of lockdown, and announcement of measures aimed at protecting businesses and jobs.
The International Labour Organization (ILO) definition of employment includes those who worked in a job for at least one hour and those temporarily absent from a job. Workers furloughed under the Coronavirus Job Retention Scheme (CJRS), or who are self-employed but temporarily not in work, have a reasonable expectation of returning to their jobs after a temporary period of absence. Therefore, they are classified as employed under the ILO definition.
The single-month estimate for the employment rate in the UK, for May 2020, shows an increase of 0.5 percentage points compared with the previous month.
The single-month estimate for the unemployment rate in the UK, for May, shows an increase of 0.2 percentage points compared with the previous month.
The single-month estimate for the economic inactivity rate in the UK, for May, shows a decrease of 0.7 percentage points compared with the previous month.
The weekly estimates show an increase in average actual hours through May, with self-employed hours rising sharply to come back in line with employees.
The weekly estimates show that approximately half a million employees in the UK were receiving no pay while their job was on hold and affected by the coronavirus pandemic.
The weekly estimates of those starting new jobs has decreased since lockdown began on 23 March 2020.
The single-month estimate of the employment rate, for people aged 16 to 64 years in the UK, for May 2020, was 76.5%. This represents an increase of 0.5 percentage points compared with the previous month (April 2020), but the rate was largely unchanged compared with three months ago (February 2020). The headline estimate for the three months March to May 2020 decreased by 0.2 percentage points compared with the previous quarter (December to February 2020) and stood at 76.4%.
The single-month estimate for the unemployment rate, for people aged 16 years and over in the UK, for May 2020, was 4.1%. This represents an increase of 0.2 percentage points compared with the previous month (April 2020) and an increase of 0.1 percentage points compared with three months ago (February 2020). The headline estimate for the three months March to May 2020 was largely unchanged on the previous quarter (December to February 2020) and stood at 3.9%.
The single-month estimate for the economic inactivity rate, for people aged 16 to 64 years in the UK, for May 2020, was 20.2%. This represents a decrease of 0.7 percentage points on the previous month (April 2020) and a decrease of 0.1 percentage points compared with three months ago (February 2020). The headline estimate for the three months March to May 2020 increased by 0.2 percentage points on the previous quarter (December to February 2020), to 20.4%.Back to table of contents
From the way the Labour Force Survey (LFS) data are collected, it is possible to separate out responses relating to individual weeks during the survey period. The Office for National Statistics (ONS) has developed a method for weighting the weekly LFS data to produce UK aggregates. The sample for any week is not representative, and the results are more volatile than the quarterly or monthly estimates. As such, their use is to show any large impact of a sudden change in labour market conditions and should not be used as a leading indicator. While not providing robust estimates of labour market conditions, these can help users to understand the impact of the coronavirus (COVID-19) pandemic on a week-by-week basis during the quarter. They may have the potential to pick up large changes in the labour market, which is why the data have been explored and are now being made available during the coronavirus period.
The weekly LFS data from 2008 to the end of May 2020 will be published monthly in this single-month article and Table X07; their usefulness and future publication will be reviewed.
Because of the experimental nature of these data, each month the seasonal adjustment is reviewed and refined to incorporate international best practice on high frequency data and better account for the specifics of the LFS sample design.
Employment and unemployment
The weekly employment and unemployment rates are extremely volatile and therefore should not be used as a leading indicator of changes to the labour market but may have the potential to pick up large changes (see Figures 4 and 5). No major change to the employment or unemployment rates are observed in any of the weeks since lockdown (week 13 in Quarter 1 (Jan to Mar)).
Lockdown measures were introduced on 23 March 2020 (during week 13 of Quarter 1), when we saw the largest falls in average actual hours (see Figure 6). This change is observed in both the employees and self-employed data, with the largest decrease seen for those identifying as self-employed. During May, we saw average actual hours start to increase for both groups, with self-employed hours increasing the most and almost coming back in-line with the employee estimates.
The LFS also collects information on reasons for why people have worked fewer hours in the reference week. Since lockdown was announced, working fewer hours because of “Economic conditions” saw large increases (this category is where LFS interviewers were advised to code those on the Coronavirus Job Retention Scheme (CJRS)). The estimated number in this category rose from fewer than 100,000 people in the long-term trend to over 5 million people by the end of April. Although the level was still over 4 million in May, there is some evidence of a decrease throughout the month.
Those temporarily away from a job
The LFS collects information on those temporarily away from a job that they expect to return to. Figure 7 shows a large increase, in both March and April 2020 (not seen in the previous three years), in those stating that they are temporarily away from paid work. Throughout May 2020, those temporarily away from a job started to slowly decrease.
Those starting a new job
The LFS collects information on the month and year that a respondent started working with their current employer or started as continuously self employed. From these responses, an estimate for the number of persons starting a new job in the reference month or previous month can be calculated. Figure 8 shows a decrease in the number of new starters since March.
Those still being paid while their job is on hold or affected by the coronavirus
In April 2020, a number of questions were added to the LFS questionnaire to gather additional information on the situation in the labour market during the coronavirus pandemic. One such question asks whether an employee is still being paid while their job is on hold and affected by the coronavirus pandemic; therefore, everyone answering this question will be defined as in employment.
Figure 9 shows approximately half a million employees receiving no pay while their job was on hold or affected by the coronavirus pandemic. Those receiving full pay peaked at just over 4 million during the first week of May and has decreased since to approximately 2.2 million in the final week of May. Those receiving partial pay fluctuated between 3 and 4 million throughout April and May.
Those who have applied for the coronavirus Self-employment Income Support Scheme
A question was also asked to those self employed who were temporarily away from work because of the coronavirus pandemic on whether they had applied for the coronavirus Self-employment Income Support Scheme (SEISS).
By the middle of May, it is estimated that over 1 million self-employed persons had applied for the coronavirus SEISS, following an increase throughout April and early May. Approximately 400,000 persons were not eligible for the scheme, a figure that remained fairly constant throughout April and May.
Back to table of contents
The Labour Force Survey (LFS) single-month estimates provide additional information about the latest quarterly movements in the headline three-monthly aggregates of employment, unemployment and economic inactivity. The production and evaluation of the estimates is an important part of our quality assurance of the three-monthly averages published in the Labour market overview.
Single-month estimates are based on one-third of the sample of the three-monthly series; this is approximately 15,000 households. Consequently, sampling variability of the changes in the single-month estimates is higher in relative terms than those of the headline aggregates, and so any interpretation of them can only be in fairly broad terms.
The LFS sample is designed so that the data collected for any three consecutive monthly reference periods (or rolling quarters) are representative of the UK population. However, the data for any given single month are unlikely to be representative of the UK. These sampling effects can cause movements in the single month that are a consequence of the survey nature of the LFS and are not a true reflection of change in the wider economy. The movement in the latest single-month figures is, in theory, a better indication of the latest change in the labour market than the difference between the latest two overlapping three-month periods, but it must still be treated with caution.
The sample design of the LFS often produces clear patterns in the single-month series, which can aid interpretation of the LFS aggregates. The estimates help users determine how closely the movements in the headline aggregates reflect changes in the UK labour market and how far they reflect the survey nature of the LFS, in particular sampling variability. For example, 80% of the households surveyed in one month will also have been surveyed three months ago. This means the comparison between the latest month and three months ago usually provides a better indicator of the latest underlying change than the comparison with the previous month, for which there is no sample overlap.
The single-month estimates are regarded as an official statistic and are not considered National Statistics because they do not have sufficient methodological robustness.
A methodological article explaining the background to the LFS single-month estimates and describing how they are calculated is available.
The model-based single-month LFS estimates are derived from a state-space model and aim to improve on the current single-month estimates. The model uses single-month wave-specific time series estimates for each variable, along with estimated variances, to produce modelled seasonally adjusted time series. These new model-based estimates are considered to be Experimental Statistics.
A methodological article giving more detail regarding the new series is available.
A new weighting methodology has been used specifically for the weekly LFS, using age, sex and region in the calibration groups.
Weekly LFS estimates have been seasonally adjusted using a modified version of TRAMO-SEATS to handle higher frequency time series. The seasonally adjusted estimates have the seasonal frequencies removed and frequencies at a 13-week lag to account for seasonality resulting from the survey design. The trend is simply a smoothed estimate of the seasonally adjusted series, the weights of which are determined by the ARIMA model and as such are "tailored" to the series.
More information can be found in the Background and methodology sheet within Table X07.Back to table of contents
Further information about the LFS is available from the LFS -- user guidance.
After EU withdrawal
As the UK leaves the EU, it is important that our statistics continue to be of high quality and are internationally comparable. During the transition period, those UK statistics that align with EU practice and rules will continue to do so in the same way as before 31 January 2020.
After the transition period, we will continue to produce our labour market statistics in line with the UK Statistics Authority's Code of Practice for Statistics and in accordance with International Labour Organization (ILO) definitions and agreed international statistical guidance.Back to table of contents
Contact details for this Article
Telephone: +44 (0)1633 455400