1. Main points

  • Early indications of the impact of the coronavirus (COVID-19) pandemic on the labour market show that average weekly actual hours worked fell by 2.5% between January and March 2019 and the same period in 2020, compared with a decline of 2.2% in the period January to March 2008 and the same period in 2009.

  • Between January to March 2019 and January to March 2020 the largest loss of average actual hours worked was recorded in the accommodation and food services industry (negative 11.8%)

  • Young workers aged 16 to 24 years experienced the largest fall in average actual hours (negative 5.9%) compared with other age groups, followed by those aged 65 years and older (negative 4.8%).

  • Vacancies decreased across all industries, with the largest percentage decrease recorded in the accommodation and food services industry (negative 41.5%) in the period between February to April 2019 and February to April 2020.

Back to table of contents

2. The coronavirus pandemic and the UK labour market

New data and indicators have shown the impact of the coronavirus (COVID-19) pandemic on the UK economy. The latest gross domestic product (GDP) estimates for Quarter 1 (Jan to Mar) 2020 showed that the UK economy shrank by 2.0%, the largest such fall since Quarter 4 (Oct to Dec) 2008. Although the latest headline labour market estimates do not yet fully show the impact of the pandemic on employment and unemployment, the estimates of vacancies and hours worked already show sharp changes. For example, the number of vacancies in the three months to April 2020 fell by 170,000 compared with the previous three months, down to 637,000.

An experimental monthly flash estimate of paid employees from Pay As You Earn (PAYE) Real Time Information (RTI) data for April 2020 also indicates that the number of paid employees fell by 1.2% compared with April 2019 and by 1.6% compared with March 2020. This is consistent with external sources; for example, the KPMG and REC, UK report on jobs highlighted that the coronavirus pandemic prompted some companies to put recruitment plans on hold. It also reported falls in vacancies for both permanent and temporary workers at the sharpest rate since the survey began in October 1997. Research produced by the Institute for Fiscal Studies (IFS) report on job vacancies during the coronavirus pandemic shows that by the time the lockdown was announced, vacancies had already decreased by around 15%.

The fall in number of hours worked since the lockdown began on 23 March 2020 is reflected in trends observed in indicators of business activity. The Office for National Statistics' (ONS') Business Impact of COVID-19 Survey (BICS) for the period 20 April to 3 May 2020 showed that 20% of the 6,196 businesses that responded had temporarily closed or paused trading. The Bank of England Monetary Policy Report for May 2020 reported the Bank of England and Ipsos MORI survey in April 2020 showed that over half of respondents reported that they were employed but not currently working or were working fewer hours as a result of the coronavirus.

This analysis adds to the latest headline labour market estimates and existing analyses of the impacts of the coronavirus pandemic on the UK economy. We will analyse how the pandemic is impacting on actual hours worked and on vacancies, and we will analyse which industries and workers were most affected by the changes in actual hours worked. We start by analysing actual hours worked followed by vacancies. We end by examining trends in experimental weekly Labour Force Survey (LFS) and monthly vacancies data.

Back to table of contents

3. The impact of the coronavirus pandemic on actual hours worked

One indicator that captures the early impact of the pandemic is the change in actual hours worked. Figure 1 shows the annual percentage changes of average weekly actual hours worked between January to March 2007 and January to March 2020. This period includes the 2008 to 2009 economic downturn, making it possible to compare that period and the current one. In the period January to March 2009, compared with the same period in 2008, average weekly actual hours worked experienced the sharpest decrease of 2.2%.

Between January to March 2020 and the same period in 2019, average weekly actual hours worked fell by 2.5% to 31.4 hours. The current drop in average weekly actual hours is larger than that experienced between January to March 2008 and the same period in 2009 (a fall of 2.2%), even though the current survey data only capture the impact of the pandemic in the last three weeks of March 2020. The full impact of the pandemic on weekly average actual hours worked will be revealed in figures for the coming months.

Back to table of contents

4. The impact of the coronavirus pandemic on hours worked by industry

The impact of the pandemic on actual hours worked varies by industry. This is because industries have different working patterns and they have been impacted by the pandemic in different ways. For example, when "stay at home" guidance was introduced, some industries had more opportunities to work from home than others, depending on a variety of factors, including differences in technology usage across industries.

Figure 2 shows the percentage change in average actual hours between January to March 2019 and January to March 2020 by industry. It shows that there was a reduction in average actual hours worked across all industries. It also shows that the largest loss of actual hours worked was in the accommodation and food services industry (negative 11.8%) and the smallest loss was in the human health and social work activities industry (negative 0.4%).

People were advised to “avoid” bars, restaurants and other indoor leisure venues on 16 March 2020 by the government, and “measures to close entertainment, hospitality and indoor leisure premises” were introduced from 20 March 2020 across the UK. As a result, the accommodation and food services industry has been among the hardest hit in the UK economy. The Business Impact of COVID-19 Survey (BICS),covering the first two weeks of lockdown (23 March to 5 April 2020), showed that across the 6,150 businesses that responded and were either continuing to trade or had temporarily paused trading, 27% of the workforce had been furloughed, with the highest furloughing rate recorded in the accommodation and food service activities (80%).

A high number of furloughed workers results in reduced hours worked, which affects the wider economy. In the gross domestic product (GDP) monthly estimate for March 2020, the accommodation and food services industry declined by 9.5% in the three months to March 2020. This was the largest decline for all industries and a major contributor to the fall in the total services industry growth.

The construction industry had the second largest reduction in average actual hours. The average actual hours worked in January to March 2020 were 5.8% fewer than in the same period in 2019. Further, the recent GDP estimates show that output in the construction industry declined by 2.6% in the three months to March 2020.

Limited opportunities for working from home in the accommodation and food services and the construction industries partially explain why they had the largest decreases in actual hours worked between January to March 2019 and January to March 2020. An Office for National Statistics (ONS) study in March 2020 on coronavirus (COVID-19) and homeworking showed that in 2019, only 10.0% of employees in the accommodation and food services industry and 25.9% in the construction industry had ever worked from home. The extent to which a worker has the ability to complete tasks from home depends on whether they are required to work in a specific physical environment, with certain tools or machinery, or in proximity to others. Technology, specifically designed to enable the completion of tasks remotely, also affects the ability to work from home.

Actual hours worked in the human health and social work activities industry had the smallest decline of 0.4% in the year to January to March 2020, followed by the finance, insurance and real estate activities industry with a decline of 1.9%. The estimates suggest that the one week of lockdown in March had little effect on actual hours worked in these sectors. The human health and social work activities industry was classified as offering essential services at the forefront of fighting the coronavirus pandemic, and many of the workers are classed as key workers. Since many of the workers have not had to stop working because ofcoronavirus restrictions, furloughing in the industry was reported to be relatively low at 11.7% in the BICS for the period 23 March to 5 April 2020.

Back to table of contents

5. The impact of the pandemic on hours worked by age

Actual hours worked differed by age group. Figure 3 shows that young workers (those aged 16 to 24 years) had the largest reduction in actual hours worked (negative 5.9%), followed by workers aged 65 years and over (negative 4.8%).

Young workers are most concentrated in the distribution (21.4%) and accommodation and food services (16.3%) industries. They also constitute a large proportion of workers on non-standard forms of employment (for example, zero-hours contracts). Older workers in the age groups 50 to 64 years and 65 years and over have higher employment concentrations in the human health and social work activities (15.8% and 12.6% respectively) and education (11.2% and 12.7% respectively) industries.

Back to table of contents

6. The change in actual hours worked by sex

The decrease in actual hours worked in the year to January to March 2020 differed between men and women. In Figure 4, the top four industries with larger decreases in actual hours worked by women than by men are accommodation and food services, construction, information and communication, and manufacturing.

In the two industries where actual hours worked decreased the most, the loss of actual hours worked by women was larger than that of men. However, overall, the actual hours worked by men decreased by more than for women. The latest estimates for the year to January to March 2020 show that actual hours worked by men decreased by 3.0% to a record low of 35.4 hours, while those for women decreased by 1.5% to 27.0 hours. Generally, men tend to work more hours than women, and women are disproportionately represented in part-time work.

Although the number of weekly actual hours worked by men declined in all industries in the year to January to March 2020, those worked by women increased in four industries (agriculture, forestry and fishing; transport and storage; human health and social work activities; and "other" services). The largest increase in average actual hours worked by women was in agriculture, forestry and fishing (8.0%) followed by transport and storage (3.1%).

We can further analyse the changes in hours by looking at the proportion of women employed in each industry compared with that of men. This shows that the increases in women’s actual hours worked in the agriculture and transport and storage industries should be treated with caution because of relatively small sample sizes within the women’s estimates, which may cause small changes in actual hours to appear to have relatively large effects.

Back to table of contents

7. The impact of the coronavirus pandemic on vacancies

The early impact of the coronavirus (COVID-19) pandemic on the labour market showed through falling numbers of vacancies, which are an early indicator of economic activity. The pandemic has had a direct effect on businesses' confidence about recruitment of new workers, resulting in fewer vacancies.

The Bank of England Monetary Policy Report for May 2020 indicates that heightened levels of uncertainty have caused hiring intentions to decline rapidly. The latest KPMG and REC, UK report on jobs survey reported a big drop in overall hiring activity in April 2020, with the sharpest decline in the demand for permanent and temporary staff since the survey began in October 1997. The latest vacancies estimates for the period February to April 2020 show a large decrease in the number of vacancies over the period.

Figure 5 shows the year-on-year percentage changes in the number of vacancies since 2007. It shows the depth of the 2008 to 2009 economic downturn when vacancies declined by 34.6% in the year to February to April 2009. In the year to February to April 2020, vacancies decreased by 24.8% to reach 637,000. This was the largest annual decrease in vacancies since April to June 2009. However, the current total number of vacancies remains higher than the post-recession low level of 432,000 recorded in the period April to June 2009.

The Office for National Statistics' (ONS) Business Impact of COVID-19 Survey (BICS) for the period 20 April to 3 May 2020 reveals why vacancies may be falling. Among the businesses that responded to the survey, 61% experienced reduced turnover, which is likely to have affected their cash flows and ability to hire new workers. However, some firms laying off staff recruited temporary staff. The IHS Markit reported a 28.8% fall in consumer spending in April 2020, which indicated falling demand for goods and services; this may have contributed to reduced demand for labour.

The decline in vacancies was recorded across firms of all sizes in the period February to April 2020, but it was more severe among smaller firms. Small firms with 1 to 9 employees and 10 to 49 employees recorded the largest decreases in vacancies on the year of 36.3% and 36.1% respectively. Very large firms, employing 2,500 workers and over, experienced the lowest decrease in vacancies of 13.0%.

The number of vacancies decreased across all industries, but the magnitude of the decrease varies across industries, as shown in Figure 6.

The largest percentage decreases were recorded in the accommodation and food services industry (41.5%) followed by “other” services (35.3%) and information and communication (34.5%); the smallest decrease was recorded in the real estate activities industry (0.8%).

Back to table of contents

8. Early indicators of weekly hours and monthly vacancies estimates from the ONS’ experimental data

The Office for National Statistics (ONS) has been producing experimental data on some labour market indicators to identify the latest trends. This analysis focuses on the experimental weekly Labour Force Survey (LFS) and monthly vacancies estimates. The experimental data should be interpreted with caution because they are highly variable and methodologically less robust compared with three-month estimates. More information on the methodology can be found in single-month LFS estimates released in May 2020. As such, we focus on scales of magnitude and directional changes rather than specific numbers.

The early impact of the coronavirus (COVID-19) pandemic showed through an increase in the number of people who are temporarily away from paid work in the period January to March 2020. These included people who were furloughed from their jobs.

The impact of the pandemic also showed through changes in actual hours worked, with large falls in average actual hours for week 12 and week 13 of Quarter 1 (Jan to Mar) 2020, which cover the week before lockdown (16 March to 22 March 2020) and the week of lockdown (23 March to 29 March 2020), respectively. In particular, the data for 2020 show a large increase in the number of people who worked fewer than usual hours for economic reasons, which includes those who reported being on the Coronavirus Job Retention Scheme (CJRS).

In Figure 7, we show trends in the difference between actual hours worked by men and women between Quarter 1 week 1 (Jan to Mar) week 1 (the first week of January) and Quarter 2 (Apr to June) week 4 (the fourth week of April) in the years 2018 to 2020. We include the first four weeks of the second quarter (for 2018 and 2019) to show past trends and the extent and direction of the current deviation.

Figure 7 shows that generally men have more actual hours worked than women, with the trend of the gap between men and women in 2018 and 2019 averaging between 8.0 and 10.0 hours. The trend of the gap in 2020 closely followed those of 2018 and 2019 until week 11 of 2020; the gap declined from week 12. This indicates that the difference between actual hours worked by men and women declined from the same period, driven by a larger decrease in actual hours worked by men than by women.

Disaggregating actual hours worked into employee and self-employed actual hours worked helps us to identify which type of worker experienced a larger decrease in hours worked because of the coronavirus pandemic. Figure 8 shows this disaggregation for the years 2018 to 2020.

In Quarter 1 2020, both self-employed and employees’ actual hours worked decreased by more than in previous years. From week 10 of Quarter 1 of 2020, self-employed hours started to deviate from the previous years’ trends. Employees’ actual hours worked also started decreasing further in week 13 of Quarter 1 2020 in response to the lockdown measures that were introduced, and the decrease in hours worked by the self-employed continued. The series in Figure 8 show high variability that is partly associated with the changing timing of Easter in different years, so caution should be used when analysing these data.

Experimental monthly vacancies estimates give a largely similar picture to that of quarterly estimates. However, the experimental estimates show different directions of changes in some industries that have not yet shown up in quarterly estimates. For example, the percentage fall in the number of vacancies in the construction and wholesale and retail industries is much higher in experimental estimates than in quarterly estimates, meaning these are the industries to watch for evolving trends.

The experimental monthly vacancies estimates show that the accommodation and food service activities industry had the largest percentage decrease in vacancies, followed by the construction and motor-trades industries. Figure 9 shows the percentage changes in vacancies by industry between April 2019 and April 2020.

Industries designated as offering essential services and employing key workers did not show large changes in the number of vacancies, for example, human health and social work activities and utilities industries.

Another set of the ONS' experimental statistics is based on the analysis of the Adzuna weekly online job adverts to derive an indicator of weekly vacancies that provides an early indication of the trend of the number of live job adverts in the UK. Indices of the Adzuna weekly vacancies and the ONS monthly vacancies estimates for the period January 2019 to May 2020 are plotted together in Figure 10.

Figure 10 shows that the experimental vacancies estimates have similar growth trends. Despite the similarity of the trends, the experimental Adzuna job adverts data and the ONS Vacancy Survey are not directly comparable because the two sources of vacancies have different definitions of what they cover. For example, the Adzuna vacancies are based on listed online job adverts, which can include multiple job opportunities within one advert and therefore do not align directly with one job vacancy. The ONS Vacancy Survey covers vacancies defined as a vacancy for which businesses are actively seeking recruits from outside their organisation. Nonetheless, the comparison of the two data sources is important as a quality check for ONS vacancies data.

Back to table of contents

9. Summary and next steps

Although the full impact of the coronavirus (COVID-19) pandemic on the UK labour market will fully reveal itself in the coming months, actual hours worked and vacancies provide early indicators of its implications. They suggest that the impact of the pandemic was felt to different extents across industries and individuals with certain demographic characteristics.

Actual hours worked by men decreased more than those worked by women. Businesses in the accommodation and food service activities industry were the worst affected in reductions of actual hours worked and vacancies. Early indications are that the young and the self-employed are being disproportionately impacted by the pandemic.

Labour market estimates covering the month of April 2020, which will be released later this month, will offer a more comprehensive picture of the impact of the coronavirus pandemic on UK workers, including on employment, unemployment and inactivity. However, some of these effects may be limited by the UK government interventions aimed at protecting workers, such as the Coronavirus Job Retention Scheme (CJRS).

Back to table of contents

Contact details for this Article

Blessing Chiripanhura
economic.advice@ons.gov.uk
Telephone: +44 (0)16335 82512