Official Statistics

Coronavirus Job Retention Scheme statistics: September 2020

Updated 8 October 2020

Overview

This is the fourth release of Official Statistics on the Coronavirus Job Retention Scheme. It provides analysis of claims for periods up to 31 July. The data used includes claims submitted to HM Revenue and Customs (HMRC) by 31 August 2020. The data for July is incomplete as claims relating to July may still be filed; thus, the figures for July should therefore be considered preliminary results and will be revised in future releases.

We will continue to publish more CJRS statistics in future months. These statistics are Experimental Statistics and are subject to revisions. Further information is provided later in this publication.

The government announced the Coronavirus Job Retention Scheme (CJRS) on 20 March 2020. CJRS supports employers through the COVID-19 period. Employers are able to claim CJRS support for employees furloughed from 1 March 2020. Up to 30 June 2020, CJRS provided employers with financial support of up to 80% of their employees’ salaries. This support is capped at £2,500 per month per employee. Employers may also claim Employer National Insurance and minimum automatic enrolment pension contributions. Starting from 1 July 2020, except in certain exceptional circumstances, staff who have not already been furloughed under the scheme may not be included in claims for support. Further changes from 1 August and 1 September – as described later in this bulletin – reduced the total level of support available for each furloughed employee.

This release extends the daily time series presented in the last release up to the end of July and includes a snapshot of furloughing by region and for flexible furloughing at 31 July.

Claims data for August 2020 is currently too incomplete to produce reliable figures. Statistics on the use of the CJRS in August and onwards will be included in future publications.

The previous CJRS statistics release published in August 2020 reported that in total, 9.6 million employments had been furloughed through CJRS (32% of eligible employments) for at least part of the period between March to June 2020. These claims were made by 1.16 million employers, with 61% of eligible employers claiming.

Contact details

For queries or feedback on this publication, please contact:

• Peter Lumb

For press queries, please contact:

• Dan Allen, tel: 03000 585 024
• Lisa Billard, tel: 07773 091 264

Next release

The next release will be published on 22 October 2020. This date has been amended from 16 October as announced on GOV.UK.

Main points

The key points from this release covering the period to 31 July 2020 are below. Figures for 31 July are provisional and subject to revision as additional claims for the period are received.

• the CJRS claims data for July is not yet complete as claims for periods from 1 July onwards may still be made - this means that the figures for July are preliminary and are likely to be revised upwards somewhat in future releases
• following on from the last release which covered the period to 30 June, preliminary estimates show that furloughing of staff across all sectors continued to decrease up to 31 July
• since the peak of 8.9 million employments furloughed on 8 May, followed by reductions in June, preliminary figures show that the number of employments furloughed continued to fall throughout July to 4.8 million on 31 July - it is estimated that this figure could be in the region of 10% higher, or around 5.3 million once all returns are received and revisions made
• furloughing of staff in the wholesale and retail sector peaked on 24 April at 1.85 million employments - by 31 July, initial figures show 789,000 jobs furloughed in this sector – a decrease of more than a half since the peak for the sector
• across all employer sizes, the number of employments furloughed continued to decrease between 30 June and 31 July
• as at 31 July, the sector with the highest proportion of its workforce eligible for furloughed that were actually furloughed was arts, entertainment and recreation at 45% followed by accommodation and food services sector at 43% - in all, 58% of employers in the arts, entertainment and recreation sector were using the furlough scheme at the end of July and 57% of employers in accommodation and food services
• overall, where it was possible to link the CJRS data on individual furloughed employments to other data, for the UK as a whole, 2.26 million women were furloughed at the end of July compared with 2.15 million men
• preliminary estimates show there was broad consistency in furlough rates across the nations and regions of the UK at the end of July - London had the highest take-up rate of 17% against the UK average of 16%
• in most countries and regions more women than men were furloughed at 31 July, the greatest exception to this being the West Midlands where 187,000 women were furloughed and 197,000 men
• almost 200,000 employers were using flexible furlough at the end of July for at least one job - in total over 950,000 employments were on flexible furlough at the end of July – about 20% of all furloughed employments
• at 29%, the accommodation and food services sector had the highest proportion of employments furloughed flexibly
• based on the preliminary data for 31 July, the South West has the highest proportion of employments on flexible furlough at 25% and London has the highest proportion of employments furloughed on a full-time basis at 82%

About the data in this release

The data used in this release comprises the CJRS claims made up to 31 August 2020. Where possible, this data has been matched with other HMRC data to enable additional analysis to be presented. This bulletin includes breakdowns of CJRS claims as follows:

• the number of employments furloughed by day, also broken down by sector and employer size

The use of CJRS as at 31 July by:

• employer size
• sector of the economy

New additions / expanded tables

• the use of flexible (partial) furlough by sector and region as at 31 July
• furloughed employments by region and gender as at 31 July

These statistics have been produced using data from both the CJRS and HMRC’s Pay As You Earn (PAYE) Real Time Information (RTI) system. Following user feedback, we continued to include details of the number of employments furloughed each day. We would welcome any further feedback from users. This should be sent to CJRS.Statistics.Enquiries@hmrc.gov.uk. User responses will then be taken into consideration when developing future releases.

For these statistics, an ‘employer’ is defined as a PAYE scheme, and ‘employments’ are defined according to the CJRS criteria; further details are within the background section.

As with previous releases, where we have been unable to match the CJRS claims with other HMRC data, we have denoted this as ‘unknown’ in the tables within this release.

The release reports a CJRS take-up rate for both employers and employments, where appropriate. The take-up rate has been calculated using PAYE RTI data, to estimate the eligible population of both employers and employments. An individual with employments with more than one employer is counted once for each employment furloughed.

Rounding has been applied to the figures in the tables accompanying this release this release, with counts rounded to the nearest 100 and amounts to the nearest million.

Employments furloughed over time

This section of the release presents the number of employments furloughed each day from the scheme’s start up to 31 July 2020. The figures for July 2020 are based on incomplete data and should be considered to be provisional and subject to change. This is because claims for staff furloughed from July onwards can still be submitted. It is estimated that the overall total number of people furloughed at the end of July will be in the region of 10% higher once all claims are submitted and revisions applied. A further consequence of it being still possible to submit claims is that the available data is not complete enough to reasonably reliably extend the daily series after the end of July.

While the job retention scheme was announced on 20 March 2020, claims for furloughed staff could be made extending back to 1 March 2020 for jobs had already been furloughed in the period 1 March to 19 March 2020; therefore the daily series included in this release start before the scheme was announced.

The data presented in this section comes from the total number of furloughed staff included in each claim. Some additional detail about furlough periods is available from the job level data supplied with larger claims (covering 100 or more furloughed jobs) and it is planned to revise the series adding this extra information in the next release of these statistics. This change to the methodology may slightly reduce the size of the step changes in the series seen between months and apparent in the charts. The noticeable changes at the start of months, for example between May and June, may be the result of the combination of the easing of coronavirus restrictions and administrative decision-making cycles for some employers that pay their staff monthly.

More details on the methodology can be found later in the bulletin.

The figures show the total number of employments furloughed by day. Some employees have moved on and off furlough over time. Because not all furloughed employments have been furloughed at the same time, the peak figures in the time series are lower than the total number of jobs furloughed at any time up to the end of June and published in the August release. That figure is 9.6 million. Additional jobs could normally not be newly furloughed after the end of June. The August release of these statistics also provides more detail on the overall total jobs that have been furloughed at any time since the start of the scheme.

Where figure 1 shows a reduction in the number of employments furloughed, it should not necessarily be inferred that employees have returned to work or alternatively that employments has ended. The data presented here should be interpreted in this context.

The key points to note from this figure are:

• initially the number of employments furloughed increased quickly following the announcement of the scheme on 20 March, reaching 4.8 million on 23 March and 6.8 million by the end of March
• following further strong growth at the start of April the number of employments furloughed increased gradually and peaked at 8.9 million on 8 May 2020
• after the early May peak, the number of employments furloughed decreased slowly before a fall of around 670,000 employments between the end of May and the start of June
• the number of employments that were furloughed then continued reduce throughout June to 6.8 million on 30 June
• preliminary figures show that the number of employments furloughed continued to decrease throughout July to 4.8 million on 31 July. As discussed above this figure will be revised in future releases. It is estimated that the figure could be in the region of 10% higher, or around 5.3 million once all returns are received and revisions made

Figure 1: Total employments furloughed, 1 March 2020 to 31 July 2020

Source: HMRC CJRS data

Employments furloughed over time by industry

This section breaks down of the overall time series included above by the employer’s industrial sector.

The key points to note from this are:

• following on from the last release which covered the period to 30 June, preliminary estimates show that furloughing of staff across all sectors continued to decrease up to 31 July
• furloughing of staff in the wholesale and retail sector peaked on 24 April at 1.85 million employments. By 1 June 2020 there had been a net decrease of 361,000 in this sector dropping further to 1.32 million employments furloughed by 30 June. By 31 July, initial figures show 789,000 jobs furloughed in this sector – a decrease of more than a half since the peak for the sector
• accommodation and food services peaked at 1.62 million employments furloughed on 10 April, with a net decline of 319,000 employments to 1.31 million furloughed by 30 June, by 31 July provisional figures show 942,000 people furloughed in the sector
• the manufacturing sector had a peak of 908,200 employments furloughed on 17 April, this reduced to 582,600 by 30 June, and a provisional total of 422,000 at the end of July
• in construction, furloughing peaked on 14 April with 721,000 employments furloughed, with this falling to 403,000 employments on 30 June. Furloughing of jobs in this sector decreased further to 277,000 (provisional figure) at the end of July. This was the largest proportionate decrease from the peak across all sectors (62%) except for energy production and supply but that sector had far fewer employments furloughed
• furloughing in arts and entertainment sector peaked later than other sectors on 15 May 2020 with 461,000 employments furloughed on that date. To 30 June there was been a net decrease of 62,000 employments furloughed in this sector and by the end of July provisional figures show 303,000 jobs furloughed – a decrease of about a third from the peak

Figure 2: Total employments furloughed by industry (millions) (largest 15 sectors), 23 March to 31 July 2020

Source: HMRC CJRS data

Employments furloughed over time by employer size

In addition to the sectoral breakdown we also provide a breakdown of the number of employments furloughed per day by employer size. The employer size has been calculated based on an estimate of the number of employments eligible to be furloughed. As with the cumulative employer size breakdown, we assume a PAYE scheme to be equivalent of an employer.

Presented in figure 3 below is a set of small charts, one for each employer size band. The underlying data for these charts can be found in the accompanying spreadsheet. The key points to note here are:

• across all employer sizes, the number of employments furloughed continued to decrease between 30 June and 31 July
• over 3.4 million employments were furloughed by large employers with 250 or more employments on 5 May (the peak for this category), this reduced to 2.8 million employments furloughed by 30 June. By 31 July, provisional figures indicate that furloughed employments among the largest employers had fallen to 1.7 million – the largest proportionate decrease across all employer size bands since the peak (51%)
• employers with 20 to 49 employments had a peak of 1.15 million employments furloughed on 17 April, compared with a peak of 796,200 for employers with 100 to 249 employments on 1 May
• employers with one employment had a peak of 295,700 employments furloughed on 20 April, compared to 219,400 employments furloughed on 30 June. A net decline of 76,300. This decreased further to 176,000 (provisional) by the end of July

Figure 3: Total employments furloughed (millions) by employer size (number of employees), 23 March to 31 July 2020

Source: HMRC CJRS data

Furloughing by Employer Size at 31 July

Where it has been possible to match CJRS data to Pay As You Earn (PAYE) Real Time Information (RTI), we have estimated the size of each employer in terms of number of employees potentially eligible for CJRS support.

For this data, we have assumed that PAYE scheme is the equivalent to an employer. For some employers, this is not an exact one-to-one equivalent. For example, some organisations operate multiple payrolls for different groups of employees and in other situations, a group of companies may pool their payrolls together under one PAYE scheme. However, in our view PAYE schemes provides a reasonable proxy for employers for the purposes of this release. The employer size has been calculated based on an estimate of the number of employments eligible to be furloughed.

The key points to note from tables 1a and 1b are, as at 31 July:

• provisional figures for 31 July show 39% of employers had staff furloughed at that date under CJRS and 16% of employments were furloughed
• employers with 20 to 49 employees were most likely to have claimed under CJRS to support the furloughing of staff, with 67% employers of this size having at least 1 employee furloughed
• employers with 250 or more employees had a provisional total of 1.67 million employments furloughed at 31 July. However, this represents a take-up rate of just 10%

Table 1a: CJRS claims by employer size as at 31 July – employer level

Employer size	Employers furloughing staff	Take-up rate	Value of claims made for periods to 31 July (£ million)
1	175,500	24%	916
2 to 4	276,800	42%	2,945
5 to 9	128,400	53%	3,087
10 to 19	76,700	56%	3,328
20 to 49	58,900	67%	4,967
50 to 99	18,300	63%	3,079
100 to 249	10,600	61%	3,485
250+	6,700	57%	12,394
Unknown	500	-	30
Total	752,300	39%	34,231

Source: HMRC CJRS and PAYE Real Time Information data

Table 1b: CJRS claims by employer size as at 31 July – employment level

Employer size	Employments furloughed	Eligible employments	Take-up rate
1	175,500	727,600	24%
2 to 4	520,600	1,682,900	31%
5 to 9	462,100	1,547,800	30%
10 to 19	469,400	1,792,700	26%
20 to 49	654,600	2,615,300	25%
50 to 99	400,200	1,977,600	20%
100 to 249	452,900	2,606,600	17%
250+	1,673,700	17,402,600	10%
Unknown	5,500	-	-
Total	4,814,500	30,353,200	16%

Source: HMRC CJRS and PAYE Real Time Information data

Note: the number of employments furloughed in the unknown size category is not directly comparable with the number of employers in the equivalent category in table 1a.

Furloughing by Sector at 31 July

This section presents analysis of CJRS claims according to the primary economic sector of employers’ activity. The take-up rate is also reported in this table for both employments and employers. This is presented in tables 2a and 2b, below. Key points from these tables are, as at 31 July:

• the sector with the highest proportion of its workforce eligible for furloughed that were actually furloughed was arts, entertainment and recreation at 45% followed by accommodation and food services sector at 43%. In all, 58% of employers in the arts, entertainment and recreation sector were using the furlough scheme at the end of July and 57% of employers in accommodation and food services
• the wholesale and retail sector was responsible for the greatest total value of claims up to the end of July at £6.6 billion. Accommodation and food services and manufacturing were the two sectors with the next highest values of claims to date by the end of August, with £5.2 billion and £4.2 billion claimed respectively)

More details on this are available in the methodology section.

Table 2a: CJRS claims by sector as at 31 July 2020 – employer level

Sector	Employers furloughing staff	Take-up rate	Value of claims made for periods to 31 July (£ million)
Agriculture, forestry and fishing	5,000	15%	109
Mining and quarrying	400	34%	69
Manufacturing	49,100	50%	4,247
Energy production and supply	400	32%	92
Water supply, sewerage and waste	2,500	45%	187
Construction	89,500	39%	3,203
Wholesale and retail; repair of motor vehicles	113,100	47%	6,570
Transportation and storage	28,000	39%	1,853
Accommodation and food services	72,200	57%	5,229
Information and communication	43,600	28%	987
Finance and insurance	10,600	31%	322
Real estate	19,900	43%	592
Professional, scientific and technical	108,500	35%	2,551
Administrative and support services	69,300	42%	3,140
Public administration and defence; social security	300	4%	70
Education	18,900	49%	1,008
Health and social work	39,100	39%	1,242
Arts, entertainment and recreation	22,900	58%	1,497
Other service activities	43,600	47%	1,030
Households	3,300	4%	34
Unknown and other	12,100	-	199
Total	752,300	39%	34,231

Source: HMRC CJRS and PAYE Real Time Information data and Inter-Departmental Business Register

Table 2b: CJRS claims by sector as at 31 July – employment level

Sector	Employments furloughed	Take-up rate
Agriculture, forestry and fishing	15,000	8%
Mining and quarrying	6,000	11%
Manufacturing	421,600	17%
Energy production and supply	5,000	4%
Water supply, sewerage and waste	18,100	10%
Construction	277,000	22%
Wholesale and retail; repair of motor vehicles	788,800	17%
Transportation and storage	208,000	16%
Accommodation and food services	942,300	43%
Information and communication	143,500	12%
Finance and insurance	45,300	4%
Real estate	76,800	18%
Professional, scientific and technical	380,200	17%
Administrative and support services	515,000	19%
Public administration and defence; social security	8,100	1%
Education	204,400	6%
Health and social work	236,300	6%
Arts, entertainment and recreation	303,000	45%
Other service activities	180,800	32%
Households	4,100	3%
Unknown and other	34,900	-
Total	4,814,500	16%

Source: HMRC CJRS and PAYE Real Time Information data and Inter-Departmental Business Register

Geography – Countries and Regions at 31 July including gender breakdown

We also provide geographic breakdown of CJRS claims based on the residential address information that HMRC holds for employees. This does not directly translate to the employee’s usual place of work, or employer’s centre of operations which may be in a different region. For example, an employee who lives in Wales and normally commutes daily to work in Bristol would be included within the count for Wales, rather than for South West England.

Figure 4 shows the number of furloughed employments by each of the English regions, and the totals for England, Wales, Scotland and Northern Ireland. For some employments it has not been possible so far to link the employment to the employees’ area of residence and these are included in the unknown category.

The key points from this figure are as at 31 July (provisional figures):

• London and the South East accounted for over 1.32 million of the furloughed employments
• the East Midlands and West Midlands had a total of 675,200 employments furloughed under CJRS
• Wales, Scotland and Northern Ireland combined had 664,900 employments claimed for under CJRS
• overall, where it was possible to link the data, across the UK as a whole, 2.26 million women were furloughed at the end of July compared with 2.15 million men. However, for 482,000 employments it had not been possible to link the furloughed employments to other HMRC data to determine the region of residence and gender of the furloughed employees
• in most countries and regions more women than men were furloughed at 31 July, the greatest exception to this begin the West Midlands where 197,000 women were furloughed and 187,000 men

Figure 4: CJRS Furloughed employments as at 31 July by region and country

Region	Female	Male	Total
London	364,200	355,600	719,800
South East	325,300	281,500	606,800
North West	227,600	226,100	453,700
East	212,000	183,200	395,200
West Midlands	186,700	197,000	383,700
South West	190,400	167,100	357,500
Yorkshire And The Humber	158,200	157,700	315,900
East Midlands	150,400	141,100	291,500
North East	71,500	71,300	142,800
Wales	100,000	95,600	195,600
Scotland	182,900	183,800	366,700
Northern Ireland	50,700	51,900	102,600

Source: HMRC CJRS and PAYE Real Time Information

Figures 5 and 6 show the take-up rate of employments furloughed within each country and region using CJRS, with figure 6 providing a gender breakdown.

The key points to note from figures 5 and 6 are as at 31 July (provisional figures):

• in figure 5, preliminary estimates show there was broad consistency in furlough rates across the nations and regions of the UK at the end of July (provisional data). London had the highest take-up rate of 17% against the UK average of 16%
• as shown in figure 6, preliminary estimates show that the take up rate is similar for men and women across all regions and countries of the UK

Figure 5: Employment furlough take-up rate at 31 July by Country and Region

Source: HMRC CJRS and PAYE Real Time Information. Office for National Statistics licensed under the Open Government Licence v.3.0. Contains OS data © Crown copyright and database right 2020

Figure 6: Employment furlough take-up rate at 31 July by Country and Region

Region	Female	Male
London	16.6%	16.6%
South East	15.1%	13.4%
North West	13.7%	14.0%
East	14.5%	12.8%
West Midlands	14.5%	15.3%
South West	14.8%	13.5%
Yorkshire And The Humber	13.1%	13.2%
East Midlands	13.6%	12.8%
North East	12.6%	13.0%
Wales	14.9%	14.9%
Scotland	14.5%	15.1%
Northern Ireland	12.3%	13.7%

Source: HMRC CJRS and PAYE Real Time Information

The data for both figures 5 and 6 can be accessed from the spreadsheet accompanying this bulletin.

Use of flexible furlough at 31 July by sector

Before 1 July, employers could only furlough staff full time. It was not possible under the scheme for an employee to, for example, work half their normal hours and for the employer to claim furlough support under CJRS for the remaining hours. From 1 July additional flexibility was introduced into the scheme whereby employees could be partially furloughed. Under this type of arrangement, the employee would work part of their normal hours for their employer and be furloughed for the rest of the time. Both part-time and full-time employees can be furloughed for part of their contracted hours under flexible furlough arrangements.

The key points to note, as per the provisional figures for 31 July:

• 199,300 employers had at least one employment on flexible furlough
• at 29%, the Accommodation and food services sector had the highest proportion of employments furloughed flexibly
• 955,200 employments were on flexible furlough: 20% of the total employments on furlough
• 3.5 million employments were furloughed on a full-time basis, 73% of the total employments on furlough. Due to missing information on some furloughed employments – for example incomplete or not fully processed spreadsheet-type returns for employers furloughing 100 or more staff – mean that whether an employee has been flexibly furloughed is not known in 7% of cases. This figure should reduce over time

Figure 7: Employees furloughed at 31 July by Sector

Source: HMRC CJRS and PAYE Real Time Information

Use of flexible furlough at 31 July by country and region

The spreadsheet file accompanying this release additionally provides figures for the proportion of the workforce furloughed in each area and sector. The analysis is based on where employees live rather than where they work.

The key points to note, as at 31 July (provisional data):

• the South East had the highest number of staff flexibly furloughed at 31 July (131,000) and the South West had the highest proportion of employments on flexible furlough at 25%. London had the lowest proportion of employments flexibly furloughed across all regions and countries of the UK at 18%

Figure 8: Employees partially and fully furloughed by Country and Region – as at 31 July

Region	Employments fully furloughed	Employments partially furloughed	Total
London	590,800	128,900	719,700
South East	475,900	130,900	606,800
North West	354,000	99,700	453,700
East	310,100	85,000	395,100
West Midlands	298,100	85,600	383,700
South West	266,900	90,600	357,500
Yorkshire And The Humber	243,300	72,500	315,800
East Midlands	227,800	63,800	291,600
North East	111,200	31,700	142,900
Wales	153,300	42,300	195,600
Scotland	288,400	78,400	366,800
Northern Ireland	78,100	24,500	102,600

Source: HMRC CJRS and PAYE Real Time Information

Background

CJRS has been introduced by the government to support employers through the COVID-19 period, this has commonly been referred to as the furlough scheme. 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). Up to the end of July, the scheme also met some of the cost of employer pension contributions and the employer National Insurance Contributions.

The scheme is based around HMRC’s Pay As You Earn (PAYE) system. For an employer to qualify for the scheme they need to have created a PAYE scheme by 19 March 2020. In these statistics, an employer is defined as a PAYE scheme. The rules for an employment to qualify to be covered by the scheme are set out in guidance, and two of the key rules are that the furloughed employee must have been employed on 19 March 2020 and the employer must have submitted a Real Time Information (RTI) submission to HMRC for the employee by this date. Further information on the qualifying criteria are available on the GOV.UK website.

Changes to the scheme from 1 July

From 1 July 2020, employers have the flexibility to bring furloughed employees back to work part time. Additionally, employers have the flexibility to decide the hours and shift patterns of their employees – with the government continuing to pay 80% of salaries for the hours they do not work. From 1 September 2020, the scheme has supported 70% of salaries for the hours for hours not worked. When claiming the CJRS grant for furloughed hours, employers will need to report and claim for a minimum period of a week.

The scheme closed to new entrants from 30 June 2020. After this date, employers have only been able to furlough employees they furloughed for a full three-week period prior to 30 June 2020.

As a consequence of the closure of the scheme to additional employees and the minimum three-week furlough period that applied until the end of June, the final date that an employer could have furloughed an employee for the first time was 10 June 2020. Employers had until 31 July 2020 to make any claims in respect of the period to 30 June 2020. A small number of claims have been made since this date. These claims relate to exceptions as set out in the CJRS guidance (for example, for employees being furloughed who have been absent from work and who had been paid Statutory Maternity Pay).

Glossary

An employer is defined within this release as a Pay As You Earn (PAYE) Scheme. In some circumstances this does not map directly to what is commonly understood to be an employer. For example, some organisations operate multiple payrolls, and in other situations, a group of companies may pool their payrolls together under one PAYE scheme. However, in our view PAYE schemes provides a reasonable proxy for employers for the purposes of this release.

An employment is defined within this release as anyone who meets the scheme criteria set out within the published guidance. We have applied this definition in order to keep the presentation simple and the difficulty in separating out Office Holders (such as directors) from what are generally considered employees. Further information on the eligibility criteria is set out in the background section above.

Measuring the data

Data source and collection

The data for this release comes from HM Revenue and Customs’ CJRS claims. It covers the whole population rather than a sample of people or companies, and it will allow for more detailed estimates of the population.

The release is classed as Experimental Statistics as the methodologies used to produce the statistics are still in their development phase. As a result, the series are subject to revisions.

Additional data from HMRC’s Real Time Information system has been matched with CJRS data in order to produce the statistics released here.

Coverage

This publication covers all CJRS claims made by employers from the start of the scheme up to 31 August 2020 for support for the wages of furloughed staff up to 31 July 2020.

Upcoming changes

Future bulletins are planned to include additional statistics. The further analysis will be informed by user feedback. Please email CJRS.Statistics.Enquiries@hmrc.gov.uk if you would like to offer feedback on how the contents can be improved in the future.

Methodology

This section provides notes on the methodology and implications for interpreting the figures.

The statistics in this release count employments. Therefore, an employee with jobs at two employers will be counted twice if both jobs are furloughed.

The take-up figures and the breakdown by employer size are based on a list of employees employed on 19 March 2020 and included in PAYE Real Time Information submissions for the 2019 to 2020 tax year. Only employments in RTI submissions received by HMRC by 19 March 2020 are counted. Following the criteria for qualifying for the scheme, this is supplemented by a list of people who were employed on 28 February 2020 but who left their job before 19 March and who were later re-employed by the same employer.

In some circumstances HMRC holds incomplete information about employments, for example where a leaving date had not been submitted by employer. In this situation, an estimate of the probability that an individual was employed on the qualifying dates has been used. The assessment of whether a person was employed on the qualifying dates is based on the methodology used for the joint HMRC/ONS statistics release, Earnings and employment from Pay As You Earn Real Time Information.

Employers making claims for 100 or more employees are required to submit the details of the employees furloughed in a spreadsheet-type file. While these claims have been processed from a customer service perspective, the processing of this information for these statistics has been complex and the processing of data on some employments has not been completed. This is a factor behind the unknown category in the tables. The completeness of this data has gradually improved since the first release, and we expect some further (smaller) improvements.

The geographic and flexible furlough breakdowns of employments include all employees that employers have furloughed where it has been possible to link claim data on furloughed employments to classifying information (such as the employee’s gender and address). The linking has been performed using employees’ National Insurance numbers to data held within HMRC’s Pay As You Earn (PAYE) Real Time Information (RTI) system. When the employer-submitted National Insurance numbers are not of sufficient quality to be matched with other HMRC data, the employments have been categorised as ‘unknown’.

The geographic breakdowns in the tables and maps use a postcode lookup file from Office for National Statistics (ONS) data to link UK postcodes to geographic areas.

Industrial sector information is based on the Interdepartmental Business Register (IDBR) produced by the ONS. Where PAYE schemes are absent from the IDBR, we have used sector information from Companies House, linking on employer name where possible. This provides Standard Industrial Classification codes (UK SIC 2007) for employers that have made a claim. Where we have unable to determine SIC codes, we have reported the sector as ‘unknown’. The descriptions of the SIC 2007 sections used have been amended since the last release to better align with the standard SIC 2007 section descriptions as published by the Office for National Statistics (the categorisation used itself remains the standard groupings and is unchanged).

The time series in this bulletin show figures for the number of employments furloughed each day, using all claims submitted to HMRC by 31 August 2020. The closing date for claims for periods to the end of June was 31 July but claims for July onwards may be submitted until after the scheme closes at the end of October. Whilst we have been able to produce time series up to the end of July, at the time of the preparation of this bulletin the data available for August was not complete enough to produce reliable statistics. We will reassess the position for the next release.

In producing the time series statistics some challenges had to be tackled. These included dealing with data on amendments to claims and claims for overlapping periods. In addition, claims for 100 or more furloughed staff may include staff furloughed for varying periods. These factors combined with some incomplete data (as mentioned above) make counting the number of employees furloughed over time complex. The method employed is designed to generally prevent overcounting employments and may in certain circumstances very slightly undercount. Consequently, for the period to the end of June, the time series presented in this bulletin may typically slightly undercount the number of furloughed employments. Some additional detail about furlough periods is available from the job level data supplied with larger claims (covering 100 or more furloughed jobs) and it is planned to revise the series adding this extra information in the next release of these statistics. This change to the methodology may slightly reduce the size of the step changes in the series seen between months and apparent in the charts. We are working on possible refinements to address the points above and expect to complete this in time for the October publication. We will revise the figures if appropriate.

For July 2020, the claims data is incomplete, and the series should be considered to be preliminary. It is estimated that the overall total number of people furloughed at the end of July will be in the region of 10% higher once all claims are submitted and revisions applied.

Questions and feedback on these statistics will be welcomed and can be sent to CJRS.statistics.enquiries@hmrc.gov.uk.

Strengths and limitations

HM Revenue and Customs (HMRC) grants pre-release access to Official Statistics publications, and in accordance with the HMRC policy, pre-release access has been granted to a number of people to enable the preparation of a ministerial briefing. Further details, including a list of those granted access, can be found on HMRC’s website.

Experimental Statistics status

The release is classed as Experimental Statistics as the methodologies used to produce the statistics are still in their development phase. This does not mean that the statistics are of low quality, but it does signify that the statistics are new and still being developed. As the methodologies are refined and improved, there may be revisions to these statistics.

Rather than waiting until the development work has been completed, the statistics are being published now to involve potential users in developing the statistics. We hope that this encourages users to provide us with their thoughts and suggestions of how useful the statistics are and what can be done to improve them. Comments can be sent by email to CJRS.Statistics.Enquiries@hmrc.gov.uk.

More information about what it means for Official Statistics to be classified as Experimental Statistics is available from the Office for Statistics Regulation.

Office for Statistics regulation review

These statistics have been produced quickly in response to developing world events. The Office for Statistics Regulation, on behalf of the UK Statistics Authority, has reviewed them against several key aspects of the Code of Practice for Statistics and regards them as consistent with the Code’s pillars of Trustworthiness, Quality and Value.

Strengths of the data

The data used in this release includes all claims made by employers up until 31 August 2020 and thus covers the complete employer population having made at least one CJRS claim. We have linked CJRS data to Pay As You Earn Real Time Information data (PAYE RTI) to provide the additional information presented in this release.

Revisions

The figures in this release incorporate data on claims received by HMRC up to 31 August and cover employments furloughed up to 30 July 2020. The figures presented for July will be revised to reflect additional claims for July made during September. In future iterations of this statistics release there will be some further revisions reflecting improvements to the data processing and methodology.

Related links

Earnings and employment from Pay As You Earn Real Time Information, UK: September 2020 - Employee and earnings statistics from PAYE Real Time Information.

Labour market in the regions of the UK: September 2020 - Regional breakdowns of changes in UK employment, unemployment and economic activity

Employment in the UK: September 2020 - Estimates of employment, unemployment and economic inactivity for the UK.

Average weekly earnings in Great Britain: September 2020 - Estimates of growth in earnings for employees before tax and other deductions from pay.

HMRC coronavirus (COVID-19) statistics - Collection of HMRC data regarding COVID-19 response initiatives and policy.