1. Main points

  • One in four people who have been employees during the coronavirus (COVID-19) pandemic had been on furlough at some point between March 2020 and June 2021.

  • Employees with GCSEs as their highest qualification were more likely to have been furloughed than those with degrees or equivalent qualifications.

  • 8% of people who have ever been furloughed were no longer employed in the three months to June 2021; this is a similar proportion to employees who had never been furloughed (7%).

  • Half of those furloughed were furloughed for more than three months and this group were less likely to be employees by August 2021, when compared with those furloughed for a shorter time.

  • The skill set of people who have ever been furloughed was similar to the skill set of current employees who have never been furloughed; however, the former group identified a slightly greater number of skills they would have liked to improve for their career when compared with employees who have never been furloughed.

Back to table of contents

2. Furlough population characteristics

The Coronavirus Job Retention Scheme (CJRS), also known as the furlough scheme, was launched by the UK government to support businesses in paying their employees during the coronavirus (COVID-19) pandemic. By August 2021, the scheme had supported 1.3 million businesses and 11.6 million jobs. The scheme came to an end on 30 September 2021.

Using data from the Labour Force Survey (LFS) and Opinions and Lifestyle Survey (OPN), this article presents new analysis about all the workers in the UK who had been furloughed at any point between March 2020 and June 2021. The OPN data used is up to August 2021, and is mainly used to look at furlough duration, whether people changed employer, and the skill set of employees. Focusing on all workers ever affected by the scheme makes our analysis different to the CJRS statistics; more information can be found in the Data sources and quality section.

!

People who have ever been furloughed are defined as those that have been on furlough at any point since the pandemic started, including currently on furlough as of April to June 2021. They are compared with "employees during the pandemic, who have never been furloughed"; this is anyone who has been employed at some point since the pandemic started up to April to June 2021, but has never been on furlough, so excludes those who did not work at all during this time. When comparing groups, all differences stated are statistically significant unless stated otherwise.

According to the LFS, one in four people who have been employees during the pandemic had been on furlough at some point since March 2020.

Figure 1: Around one in four employees during the pandemic had been furloughed up to June 2021

Proportion of people who have ever been furloughed between March 2020 and June 2021, UK

Embed code

Notes:

  1. Proportion of people who have ever been furloughed refers to all those furloughed on at least one occasion between March 2020 and June 2021.

The proportion of people who have ever been furloughed was higher (30%) for workers aged under 24 years and over 65 years, compared with 23% of workers aged 35 to 44 years. Single working parents were particularly affected, as 31% of them were furloughed, compared with 24% of workers living as a couple with dependent children.

After controlling for differences in other individual and job characteristics, we find that Asian workers were 3.8 percentage points less likely to be furloughed when compared with those of White ethnicity. Although more disabled workers were furloughed than non-disabled workers (28% versus 26%), this difference was not significant once personal and job characteristics were accounted for. More individual characteristics differences in likelihoods of having ever been furloughed can be found in the associated datasets.

The likelihood of being furloughed was lower for those with qualifications above A level, even after accounting for differences in individual and industry characteristics. Employees during the pandemic with a degree or equivalent qualification were 8.9 percentage points less likely to be furloughed, when compared with those whose highest qualifications were GCSEs. This may reflect more specific job responsibilities or levels of experience across people with similar jobs, which meant certain people were less likely to go on furlough.

Current labour market characteristics

CJRS statistics up to 30 June 2021 show that workers in hospitality, construction and recreation were among the most likely to have been furloughed. LFS data also show that 55% employees who were in arts, entertainment and recreation, and 69% of employees in accommodation and food services in the three months to June 2021 had been furloughed in the past. However, Figure 3 shows that those working in the group of occupations that includes corporate managers and directors, or in professional occupations (that is, "Highest skill level") were less likely to have been furloughed in the past regardless of the industry they were in. More detailed furlough rates by occupation can be found in the associated datasets.

When controlling for individual and labour market characteristics, people working in the “highest skill level” were 8 percentage points less likely to have been furloughed than those in the “lowest skill level”.

Furlough duration

Half of people who have ever been furloughed spent more than three months in total on the scheme, across all the periods they may have been on furlough, and 24% spent six months or more furloughed.

However, some groups of people were more likely to be furloughed for longer. Data from the Opinions and Lifestyle Survey (OPN) show that, of all those who have ever been furloughed, 54% of women were furloughed for more than three months, compared with 45% of men. In addition, a greater proportion of furloughed disabled workers were furloughed for more than three months when compared with non-disabled workers (51% compared with 42%), although there could be other factors that we do not account for driving this difference.

Back to table of contents

3. Labour market flows

Of all those who had ever been furloughed, the large majority were still in work in the three months to June 2021, even if they were still on furlough. According to CJRS statistics, there were still 1.9 million employments on furlough by the end of June 2021, which represented about 16% of all employments on furlough since the beginning of the scheme. See the data sources and quality section for comparisons between the different sources, as well as the Comparison of furloughed jobs data, UK: March 2020 to June 2021 release, which compares CJRS statistics with Business Insights and Conditions Survey furlough estimates. BICS furlough estimates have been released during the pandemic by industry and region. According to the Labour Force Survey, over 9 in 10 (92%) of those who were ever furloughed were still employees, which is comparable with those who had never been furloughed (93%). The difference in employment rates between those furloughed and who had never been furloughed was not statistically significant.

!

This analysis does not measure whether people who were still on furlough by June 2021 were in work when the furlough scheme ended on 30 September 2021. The ONS will continue to monitor the outcomes of those who were furloughed as more recent data become available.

Current employees who had ever been furloughed were more likely to be in part-time work than those who were never furloughed (30% of furloughed workers compared with 23% of workers who were never furloughed). They were also more than twice as likely to be on a zero-hours contract than those who were never furloughed, at 5% and 2% respectively.

Current employees who had ever been furloughed were also more likely to be looking to change their employment situation. They were more likely to be looking for a different or additional job (7% compared with 5%) and were twice as likely to report that they were looking for this work because of the coronavirus (COVID-19) pandemic (21% compared with 12%).

Although the majority of people who have been employees during the pandemic were still employees by April to June 2021, regardless of furlough status, 8% of them were out of work.

Out of these recently out-of-work employees, people who have ever been furloughed were slightly more likely to be looking for a job and available to start, that is, unemployed as opposed to economically inactive (2.8% of people who have ever been furloughed compared with 2.1% of those who had never been furloughed were unemployed). Once differences in personal characteristics, education level and most recent job details were accounted for, people who have been furloughed were just over 1 percentage point more likely to be out of work (either inactive or unemployed).

We find that 4% of disabled furloughed people had slipped into unemployment compared with 3% of non-disabled furloughed people. In addition, 8% of disabled people who were furloughed were economically inactive (that is, not looking or available to start work) compared with 4% of non-disabled furloughed people. This might be caused by differences in their personal characteristics.

Most furloughed employees stayed with the same employer that they had initially been furloughed from. Labour Force Survey (LFS) data show that in the three months to June 2021, 86% of workers who were furloughed in April to June 2020 were still with the same employer they were furloughed from 12 months later. A higher proportion (94%) of employees who had never been furloughed remained with the same employer in the same period.

Opinions and Lifestyle Survey (OPN) data show that for those who have changed employer since being furloughed, there is no statistically significant difference between whether they found a job in the same sector (47%) or a new sector (53%). The difference between these two outcomes was not statistically significant.

Workers aged between 16 and 24 years were less likely to stay with the same employer (64%) when compared with all other age groups (25 to 44 years, 45 to 64 years and 65 to 75 years), which ranged from 79% to 91%. This continues a pattern seen before the pandemic, where young people tended to move jobs in the labour market more often than other age groups.

Longer periods of furlough were associated with a lower likelihood of being employed. OPN data show that 94% of those who were furloughed for up to three months reported being employed by July to August 2021 compared with 89% of those on furlough for more than three months. This follows previous analysis that shows the length of time out of work can affect your chances of returning to work.

Back to table of contents

4. Skills of the furlough population

People who have ever been furloughed have had a temporary period away from their job, where they have not been carrying out the roles and responsibilities their job normally demands. Previous research such as Edin and Gustavsson (2008) has suggested that prolonged absences from work may lead to workers losing some of their workplace skills. This can be detrimental as workplace skills can impact future earnings, and subsequent job satisfaction and promotion and training opportunities (see Organisation for Economic Co-operation and Development (2019) (PDF, 2.42MB)).

We used data from the Opinions and Lifestyle Survey (OPN) to understand how the skills of the people who have ever been furloughed compared with people who worked during the pandemic but werenot furloughed.  More information about the way the OPN skills data were collected and analysed can be found in the Data sources and quality section.

Despite not actively using their work skills, the skill set of the furloughed population was more like the skill set of employees, rather than the unemployed. The furloughed population reported being good at a similar number of skills, and good at the same types of skills as employed people who had never been furloughed. They also reported similar skills they would like to improve upon.

However, there were some subtle differences between people who have ever been furloughed and employees during the pandemic who have never been furloughed.

Improving skills

We asked people to report their ability against a list of 10 different skills, such as advanced maths, manual work, and presentation skills. People who have ever been furloughed identified a slightly greater number of skills they would like to improve for their career when compared with employees during the pandemic who have never been furloughed (1.8 versus 1.6).

Understanding the skills that the people who have ever been furloughed would like to improve for their career can help identify training needs for them and can also give some indication of whether they are looking to change jobs.  Figure 8 highlights some of the skills that people who have ever been furloughed were more likely to want to improve than employees during the pandemic, who have never been furloughed.

After accounting for differences in personal characteristics, education levels and occupations, those who had been furloughed mentioned slightly more skills that they would like to improve on average, at 0.1. Whether someone was furloughed or not was far less important than other characteristics such as age. On average, those aged 16 to 24 years mentioned 0.9 more skills they would like to improve on than those aged between 45 and 54 years, after controlling for other characteristics.

!

We do not know about the skills of the furloughed and the non-furloughed populations before the furlough scheme started, so we cannot say that being furloughed has caused some of these differences, as they may have already existed before the scheme began. We can say that there are small but significant differences between those who have and have not been furloughed, even after we control for differences in occupations and other characteristics.

The top three skills the people who have ever been furloughed wanted to improve for their career were:

  • advanced IT (36%)

  • creative skills (29%)

  • advanced maths (21%)

These three skills are amongst the skills that they used the least in their current or most recent job.

There are different reasons why furloughed workers may want to improve on skills for their career they did not use in the workplace. For example, a lower perceived skill level by a worker could lead to that skill being used less in the workplace. Alternatively, this could suggest that those who had been furloughed are looking to expand their current skillset to move jobs, whether that be a promotion or a different career. 

Skills workers were good at

The top three skills people who have ever been furloughed thought they were good at were the same as employees during the pandemic who have never furloughed, as shown by Figure 9. This was true regardless of the occupation people were in, and whether this occupation had high rates of furlough.

Those who have ever been furloughed reported being good at slightly fewer skills compared with employees who have never been furloughed. Among furloughed workers, 56% chose four or more skills they thought they were good at, compared with 60% of those currently employed who have never been furloughed. The figure for the unemployed who had never been furloughed was lower at 46%.

After accounting for differences in personal characteristics, education levels and occupations, people who have ever been furloughed mentioned 0.1 fewer skills that they thought they were good at on average when compared with employees during the pandemic who have never been furloughed. This is not as big as other effects we observe, such as those with a degree-level highest qualification reporting 0.3 more skills they thought they were good at on average than those whose highest qualification was GCSEs graded A* to C. However, it does suggest that people who have ever been furloughed think they are good at slightly less skills than employees during the pandemic who have never been furloughed.

Back to table of contents

5. Furlough data

Skill set of people who have been furloughed in Great Britain
Dataset | Released 1 October 2021
Skill set of people who have been furloughed: the types of skills furloughed workers think they are good at, would like to improve for their career and use in their current job.

Labour market flows for people who have been furloughed in the UK
Dataset | Released 1 October 2021
Labour market flows of people who have been furloughed.

Characteristics of people who have been furloughed in the UK
Dataset | Released 1 October 2021
Individual and labour market characteristics of people who have been furloughed.

Back to table of contents

6. Glossary

Furlough

In this article, furlough is defined as a temporary absence from work allowing workers to keep their job while the coronavirus (COVID-19) pandemic continues.

This article focuses on the population that have been furloughed at any point since March 2020. In both the Labour Force Survey and Opinions and Lifestyle Survey, respondents were asked:

"Have you been placed on furlough at any point since March 2020?"

Anyone who answers yes to this question, and is either a current employee, or was employed at some point between March 2020 and June 2021, is part of the "people who have ever been furloughed" population; that is, they have been furloughed at some point. Please note that the furlough status is self-reported.

Because of data quality reasons, we have excluded workers who were self-employed but also reported that they had been furloughed.

Furlough duration

Respondents who identified as being furloughed at any point were asked the following question on the Opinions and Lifestyle Survey:

Since March 2020, how long were you working less hours because you were on furlough?

  • two weeks or less

  • more than two weeks but less than a month

  • between one and three months

  • more than three months but less than six months

  • six months or more

After the first week of collection, an additional response option of 12 months or more was added, but in all outputs, this has been grouped with those between 6 and 12 months.

Pandemic period

In this article, we use the phrase "during the pandemic", or refer to the "pandemic period". When using Labour Force Survey (LFS) data, the "pandemic period" refers to the period March 2020 to June 2021. When using the Opinions and Lifestyle Survey (OPN) data, the pandemic period refers to the period March 2020 to August 2021. More information about the time period of the LFS and OPN can be found in the Data sources section.

Disability status

The Government Statistical Service (GSS) harmonised "core" definition identifies "disabled" as a person who has a physical or mental health condition or illness that has lasted or is expected to last 12 months or more that reduces their ability to carry out day-to-day activities. The GSS harmonised questions are asked of the respondent in the survey, meaning that disability status is self-reported.

Coronavirus Job Retention Scheme

The government announced the Coronavirus Job Retention Scheme (CJRS) on 20 March 2020. It was introduced to support employers through the COVID-19 period; this has commonly been referred to as the furlough scheme.

The scheme is based around HMRC's Pay As You Earn (PAYE) system. It works by providing grants to employers of up to a maximum 80% of salary to a maximum value of £2,500 per employee (until the end of August 2020). Up to the end of July 2020, the scheme also met some of the cost of employer pension contributions and employer National Insurance contributions.

Statistical significance

This article only presents results that are statistically significant, unless stated otherwise. Statistical significance at the 5 percentile level is used from models, while for summary statistics, non-overlapping confidence intervals were used to determine if values were significantly different from each other. More information about statistical significance can be found on our Uncertainty pages.

Back to table of contents

7. Data sources and quality

Labour Force Survey data

Sections on characteristics and flows in this article uses data collected through the Office for National Statistics' (ONS) Labour Force Survey (LFS), using the April to June 2021 data.

The LFS is a large representative survey of households in the UK. The analysis presented in this article focuses on those currently an employee or those who have had a job at some point since March 2020.

The LFS performance and quality monitoring reports provide data on response rates and other quality measures.

Opinions and Lifestyle Survey data

Sections on skills, furlough duration and whether people have stayed with the same employer uses data collected through the Office for National Statistics' (ONS) Opinions and Lifestyle Survey (OPN).

The OPN analysis presented in this article is based on pooled datasets.

Estimates for furlough duration and skills use a six-week pooled dataset, covering weekly collection from 14 July to 17 July to 18 August to 22 August 2021. After restricting the dataset to those who reported being an employee at the time of data collection, or who had a job at some point since March 2020, the sample size included 9,737 adults aged 16 years and over in Great Britain. Pooling six waves of data together increases sample sizes, allowing us to explore skills levels for different groups of the population.

Estimates for whether those who have been furloughed and employed remained in the job they were furloughed from, and whether those who have changed jobs changed sectors, uses a two-week pooled dataset, covering 11 August to 15 August and 18 August to 22 August 2021. After restricting the dataset to those who reported being an employee at the time of data collection, or who had a job at some point since March 2020, the sample size was 2,990 adults aged 16 years and over in Great Britain.

Survey weights were applied to make estimates representative of the population for both sets of pooled data (based on May 2021 population estimates).

Further information on the survey design and quality of the OPN can be found in the Opinions and Lifestyle Survey QMI.

To measure the skill set of respondents, they were asked the following questions:

(a) In which of the following do you feel competent?

Please select all that apply

  • basic IT, such as sending emails or using the internet

  • advanced IT, such as data science and programming

  • basic maths, such as price sums and percentages

  • advanced maths, such as accountancy or statistics

  • manual work, such as repairs and construction

  • using technical equipment, such as operating machines

  • customer service or sales

  • presenting or negotiating

  • creative skills, such as performance, web design or product design

  • other, please specify

(b) Thinking about your current or most recent job, which of the following do you do?

Same answer options as above

(c ) Which, if any, of these skills would you like to improve on for your career if you were offered the training?

Same answer options as above

Coronavirus Job Retention Scheme statistics

The Coronavirus Job Retention Scheme (CJRS) statistics use data from HM Revenue and Customs' claims data to measure the number of employments supported by the CJRS scheme. This means that if a person is furloughed from two jobs, this person would be counted twice in the CJRS statistics.

This is different to the analysis in this article. This article uses survey data from people who self-reported being furloughed, regardless of the number of jobs they may have been furloughed from. If a person was furloughed from two jobs, this would count only once in our analysis.

Despite the differences in concepts, there are similar distributional findings such as the differences in furlough rates for industries and age groups. However, CJRS statistics show higher levels of employments on furlough than number of people on furlough from the LFS, for April to June 2021.

Back to table of contents

Contact details for this Article

Robynne Davies, Annemari De Silva, Gueorguie Vassilev, Lualhati Santiago, James Probert
economic.wellbeing@ons.gov.uk
Telephone: +44 1633 455142