Labour Market Statistics, June 2020...Labour Market Statistics, May 2020: Claimant count analysis number of hours. While there does appear to be a discrepancy between the claimant

© Institute for Employment Studies 1

Institute for Employment Studies 1

Labour Market Statistics, June 2020 16 June 2020

This briefing note sets out analysis of the Labour Market Statistics published this morning.

The analysis covers:

■ Claimant unemployment. The ‘Claimant Count’ is a measure of the number of people

claiming benefits principally for the reason of being unemployed, compiled from

Jobseeker’s Allowance and Universal Credit data. This is not an official measure of

unemployment, but it generally follows the same trends as the official survey-based

unemployment measure and provides more timely and granular data. Today’s data

sets out claimant unemployment as at 14 May 2020.

■ Labour Force Survey data. This is a household survey that collects official data on

employment, unemployment and economic inactivity. Today’s data covers the period

February to April 2020, drawing on approximately 33 thousand interviews. Around one

third of these will have been conducted in April, so this is the first release with a full

month of data from after the crisis started.

■ Pay As You Earn Real Time Information. These are experimental statistics on

employee levels and pay, covering the period to May 2020. The statistics are drawn

from data supplied by employers in PAYE returns. Where data is missing (which

affects approximately 10% of cases in the most recent month being reported) values

are imputed based on previous returns.

Claimant unemployment is rising faster than at any point since unemployment benefits were introduced

Between March and May, claimant unemployment rose by 1.56 million, from 1.24 million

to 2.80 million – an increase of 125%. Claimant unemployment has risen more in two

months than it did in the whole of the last two recessions, and by more than it did in the

first twelve months of the Great Depression (when it rose by 1.0 million). As Figure 1

below shows, unemployment is now as high as it was in October 1993, while Figure 2

shows that this is now the biggest year-on-year rise in unemployment benefits since their

introduction in 1922.

It is important to note The Office for National Statistics has emphasised this morning that

the claimant count may be over-stating the true rise in unemployment, as some of those

being treated as unemployed in the benefits system may in fact still be working a small

http://www.employment-studies.co.uk/

Labour Market Statistics, May 2020: Claimant count analysis

number of hours. While there does appear to be a discrepancy between the claimant

count and other measures, our view is that short-hours working cannot explain most of

the rise that has been observed in the last two months. This is explored in more detail

later in the report.

Figure 1: Administrative and Claimant Unemployment, 1922-present

Source: IES analysis of Bank of England and Office for National Statistics data. The figure presents

Administrative Unemployment for 1922-1971, and the Claimant Count for 1971 onwards.

Figure 2: Year-on-year increase in claimant unemployment, 1922-present

Source: IES analysis of Bank of England and Office for National Statistics data. The figure presents

Administrative Unemployment for 1922-1971, and the Claimant Count for 1971 onwards.

Claimant unemployment is highest for young people, but is rising across all age groups

As Figure 3 below sets out, the claimant count has doubled across all age groups. It has

risen by 122% for those aged 25-34, although as we set out last month the large rise in

unemployment for older people – up from 300 thousand to 600 thousand – is very

-

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

Dec-2

2

Mar-

26

Jun-2

9

Sep-3

2

Dec-3

5

Mar-

39

Jun-4

2

Sep-4

5

Dec-4

8

Mar-

52

Jun-5

5

Sep-5

8

Dec-6

1

Mar-

65

Jun-6

8

Sep-7

1

Dec-7

4

Mar-

78

Jun-8

1

Sep-8

4

Dec-8

7

Mar-

91

Jun-9

4

Sep-9

7

Dec-0

0

Mar-

04

Jun-0

7

Sep-1

0

Dec-1

3

Mar-

17

-2,000,000

-1,500,000

-1,000,000

-500,000

0

500,000

1,000,000

1,500,000

2,000,000

Nov-2

3

Oct-

26

Sep-2

9

Aug-3

2

Jul-35

Jun-3

8

May-4

1

Apr-

44

Mar-

47

Feb-5

0

Jan-5

3

Dec-5

5

Nov-5

8

Oct-

61

Sep-6

4

Aug-6

7

Jul-70

Jun-7

3

May-7

6

Apr-

79

Mar-

82

Feb-8

5

Jan-8

8

Dec-9

0

Nov-9

3

Oct-

96

Sep-9

9

Aug-0

2

Jul-05

Jun-0

8

May-1

1

Apr-

14

Mar-

17

Feb-2

0



concerning, as previous recessions have shown that older people are much less likely to

leave unemployment and get back to work than others.

Figure 3: Claimant unemployment by age, March-May 2020

Source: IES analysis of NOMIS claimant count data

Looking at the proportion of the workforce unemployed, in Figure 41 below, we can see

that claimant unemployment is highest for young people – now standing at 12.8% of the

18-24 workforce, a rise of 6.7 percentage points since the crisis began. Given that

around half of young unemployed people do not claim benefits (usually because they are

not eligible, due to their age or studying) it is likely that the official measure of

unemployment could be on its way towards 25% in the coming months, or one million.

This would be a higher rate than was reached in the aftermath of the 2008/9 recession.

1These figures should be treated with caution, as they use estimates of the economically active population

between February and April 2020 which are likely to over-state economic activity and so under-estimate the

unemployment rate. This is because some of those who left employment since March 2020 will have

become economically ‘inactive’ rather than unemployed.

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

800,000

900,000

18-24 25-34 35-49 50+

Mar-20 Apr-20 May-20


Figure 4: Estimated claimant unemployment rate by age, March-May 2020

Source: IES analysis of NOMIS claimant count and ONS Labour Force Survey data

Claimant unemployment is highest in many ex-industrial, inner city and coastal areas

Figure 5 below illustrates that claimant unemployment rates are highest in the North East

(10.1%), West Midlands and North West (both 9.0%). The percentage point rises in

unemployment however are now fairly consistent across regions, rising by around four

percentage points since March. The exception to this is London, where the rate has risen

by 5.1 percentage points – going from having one of the lowest rates to being among the

highest.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

18-24 25-34 35-49 50+

Mar-20 Apr-20 May-20 Change Mar-May (percentage point)



Figure 5: Claimant unemployment rate by region or nation, March-April 2020


The fact that unemployment rates have increased by similar amounts between regions

also means that the percentage rise has been greater in areas where unemployment was

previously lower. So as Figure 6 shows, claimant unemployment has risen by over 170%

in the South East and South West, while it is up by 72% in the North East (albeit from a

far higher pre-crisis rate than elsewhere).

Figure 6: Percentage increases in claimant unemployment by region, Mar-May 2020


0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0%

South West

East

South East

Northern Ireland

East Midlands

Scotland

United Kingdom

Wales

London

Yorks and Humber

North West

West Midlands

North East

Mar-20 May-20 Change Mar-May (percentage point)

0%

20%

40%

60%

80%

100%

120%

140%

160%

180%

200%


Figure 7 looks beneath the regional level at individual Local Authorities, grouping them

according to the decile of unemployment immediately before the crisis began. Those

areas in the first decile had the lowest unemployment (averaging just 1.3% of residents),

while those in the tenth had the highest unemployment (averaging 5.3% of residents).

This analysis shows that the rises in claimant unemployment rates have generally been

greater in areas that had higher unemployment before the crisis began. Places in the

eighth to tenth deciles saw increases of 3.7 to 3.8 percentage points. However there

have been very large rises in areas that had fairly average levels of employment before

the crisis began, emphasising the broad based nature of the labour market impact so far.

Areas with lower unemployment have seen smaller increases in unemployment rates,

although as with the regional analysis above this disguises large percentage increases –

with unemployment more than trebling in the lowest decile, albeit from a low base.

Figure 7: Percentage point change in proportion of residents claimant unemployed, by

decile of unemployment in March 2020 (1=lowest)


Looking at individual local authorities, those areas with the highest unemployment are

dominated by ex-industrial areas, coastal towns and inner cities. Table 1 below illustrates

this, showing areas where more at least one in eleven working age adults (9%) are

unemployed.

It is worth noting that several London boroughs now feature among the local authorities

with the highest rates, which have all seen unemployment more than double since March.

The highest pre-crisis rate among these boroughs of 4.4% in Barking and Dagenham put

it 39th in the list of all local authorities sorted from the highest to lowest claimant rate.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

1 2 3 4 5 6 7 8 9 10



Table 1: Local authorities where at least one in eleven working age adults are claimant

unemployed

Local authority Region Proportion of 16-64 population % increase

May 2020

March 2020

Change (percentage point)

In claimants

Blackpool North West 12.3 7.2 5.1 72

Thanet South East 10.8 5.7 5.1 91

Birmingham West Midlands 10.6 6.7 3.9 57

Wolverhampton West Midlands 10.3 6.4 3.9 62

Middlesbrough North East 10.1 6.2 3.9 62

Haringey London 10.1 3.8 5.7 165

Barking and Dagenham London 10.1 4.4 6.3 133

Kingston upon Hull, City of Yorks and Humber 9.7 5.8 3.9 66

South Tyneside North East 9.5 6.2 3.3 53

Oldham North West 9.5 5.2 4.3 83

Hartlepool North East 9.4 6.0 3.4 56

Newham London 9.4 3.3 5.9 185

Waltham Forest London 9.4 3.5 6.1 173

Brent London 9.3 3.6 5.7 161

Bradford Yorks and Humber 9.1 5.2 3.8 75

Sandwell West Midlands 9.1 5.3 3.9 72

Hastings South East 9.1 4.9 4.2 87

Burnley North West 9.0 5.4 3.6 66

Knowsley North West 9.0 4.8 4.2 88

Lewisham London 9.0 4.0 5.0 128


Vacancies have more than halved, with eight claimant unemployed people chasing every job opening

The ONS Vacancy Survey reports that there were on average 476 thousand vacancies

over the three months to May 2020. The single month estimate of vacancies for May

however was 320 thousand. Our analysis of Adzuna online vacancy data for Joseph

Rowntree Foundation suggests that June vacancies will be around 350 thousand,

meaning that by next month the three month average being reported by the ONS will be

around 330 thousand. As Figure 8 below shows, this will be the lowest level of vacancies

in the twenty years that the survey has been running (although longer-run data suggests

that vacancies were lower in the 1990s).


Figure 8: Estimated number of vacancies

Source: IES analysis of ONS Vacancy Survey. Dotted line indicates IES estimates based on analysis of

online vacancy data supplied by Adzuna for Joseph Rowntree Foundation

Record low vacancies combined with very high claimant unemployment means that there

are now 8.5 claimant unemployed people chasing every job advert. This has risen from

just 1.5 claimants per vacancy two months ago and is already well over double the rates

in the last downturn. This is shown in Figure 9 below. Our separate analysis of local

trends in this data, which we will be updating later this week, suggests that in ex-

industrial, coastal and inner city areas these ratios are significantly higher still.

Figure 9: Number of claimant unemployed per job vacancy

Source: IES analysis of NOMIS claimant count and ONS Vacancy Survey. Dotted line indicates IES

estimates based on analysis of online vacancy data supplied by Adzuna for Joseph Rowntree Foundation.

-

100

200

300

400

500

600

700

800

900M

ay-0

1

Jan-0

2

Sep-0

2

May-0

3

Jan-0

4

Sep-0

4

May-0

5

Jan-0

6

Sep-0

6

May-0

7

Jan-0

8

Sep-0

8

May-0

9

Jan-1

0

Sep-1

0

May-1

1

Jan-1

2

Sep-1

2

May-1

3

Jan-1

4

Sep-1

4

May-1

5

Jan-1

6

Sep-1

6

May-1

7

Jan-1

8

Sep-1

8

May-1

9

Jan-2

0

Thousands

Axis Title

-

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

May-0

1

Dec-0

1

Jul-02

Feb-0

3

Sep-0

3

Apr-

04

No

v-0

4

Jun-0

5

Jan-0

6

Aug-0

6

Mar-

07

Oct-

07

May-0

8

Dec-0

8

Jul-09

Feb-1

0

Sep-1

0

Apr-

11

Nov-1

1

Jun-1

2

Jan-1

3

Aug-1

3

Mar-

14

Oct-

14

May-1

5

Dec-1

5

Jul-16

Feb-1

7

Sep-1

7

Apr-

18

Nov-1

8

Jun-1

9

Jan-2

0



Real-time Pay as You Earn data shows that the number of employees has fallen by at least 600 thousand

Figure 10 below shows the latest estimates of employee numbers from PAYE data up

until May 2020. This covers employees only, and is likely to have missing data for around

10% of staff (with these values imputed based on previous trends). This suggests that

employee numbers have fallen by approximately 600 thousand between March and May,

from 29.0 to 28.4 million.

This is a very large fall, as the figure sets out, but it is far smaller than the 1.6 million rise

in unemployment recorded in the claimant count. There are three plausible explanations

for this:

■ The employee data below is understating the fall due to incomplete returns. This

seems highly plausible, as the estimated monthly fall for previous month (March) was

revised up this morning to 450 thousand from a flash estimate of 20 thousand reported

last month. The flash estimate for April was 160 thousand, and we may well see this

also revised up next month.

■ Falls in the number self-employed. This is explored below, but it appears certain that

at least 300 thousand fewer self-employed people were in work in April than in March.

■ Claimant unemployed people are working short hours, and so not officially

unemployed. This will certainly explain part of the difference, as those claiming

Universal Credit and earning below a set threshold (roughly equivalent to sixteen hours

a week) are required to search for work and so may be counted as unemployed.

All told, our view is that the figures in the coming months will confirm that employment has

fallen by an unprecedented amount, but that the official measure of unemployment may

continue to lag the claimant count for some time.

Figure 10: Employee numbers reported through Pay As You Earn data

Source: IES analysis of Real Time Information Pay As You Earn data

25,500,000

26,000,000

26,500,000

27,000,000

27,500,000

28,000,000

28,500,000

29,000,000

29,500,000

Jul-14

Sep-1

4N

ov-1

4Jan-1

5M

ar-

15

May-1

5Jul-15

Sep-1

5N

ov-1

5Jan-1

6M

ar-

16

May-1

6Jul-16

Sep-1

6N

ov-1

6Jan-1

7M

ar-

17

May-1

7Jul-17

Sep-1

7N

ov-1

7Jan-1

8M

ar-

18

May-1

8Jul-18

Sep-1

8N

ov-1

8Jan-1

9M

ar-

19

May-1

9Jul-19

Sep-1

9N

ov-1

9Jan-2

0M

ar-

20

May-2

0


Labour Force Survey statistics are showing large rises in inactivity, but not yet unemployment

Turning to the main Labour Force Survey measures, as noted above the labour market

impacts of the crisis are not yet apparent in the headline measures of employment,

unemployment and inactivity. Overall in the three months from February to April,

employment fell by 150 thousand compared with the figures reported last month,

unemployment was broadly unchanged and economic inactivity (those not looking nor

available for work) rose by 190 thousand. However these are three month average

figures, so the single-month changes for April were larger than this.

ONS publishes single month estimates as experimental statistics, and Figure 11 below

shows the reported changes – employment appears to have fallen by just over 300

thousand while worklessness rose by the same amount.

Figure 11 Change in single-month estimates of employment, unemployment and inactivity

– March to April 2020

Source: IES analysis of Labour Force Survey data.

Looking in more detail at reasons for inactivity, it appears that the rise is more than

explained by those saying that they are inactive for ‘other’ reasons than being too sick to

work, looking after their family or home, retired, students or discouraged workers. Overall,

inactivity for these reasons actually fell in today’s data. The most plausible ‘other’ reasons

for being economically inactive would be that people were not seeking work while they

waited for their claims to Universal Credit to be processed, and/ or were self-employed

and had temporarily stopped working.

Tellingly, the number of people economically inactive who report that they ‘want a job’ has

also increased significantly compared with the data reported last month – which suggests

that these large rises in inactivity will feed through directly into increases in the headline

measure of unemployment in the months ahead.

-400

-300

-200

-100

0

100

200

300

400

Apr-20

Thousands

Employment Unemployment Inactivity



Most of the reported fall in employment is explained by fewer people self-employed

As noted, the three-month average for employment was 150 thousand lower this month

than the figure reported last month. Figure 12 shows that two thirds of this fall is

explained by lower self-employment.

Given that this is a three-month average, the implied single-month figure for April is likely

around 300 thousand lower than pre-crisis levels. It is also likely that this understates the

real fall in self-employment, as anyone reporting having spent any time working for their

business in the reference week.

Figure 12: Number of people self-employed, Mar-May 2019 to Feb-Apr 2020


While employment has not fallen, working hours appear to have collapsed

The Labour Force Survey data does however show very large falls in average working

hours in the three months to April. Figure 13 below sets these out. Overall working hours

are down by 7.3% on the figure reported last month, with average hours lower than at any

previous point. Again, because this is a three-month average, the implied single month

fall is likely to be three times this – i.e. above 20%.

This is explained by the Job Retention Scheme, with the ONS reporting that there are at

least six million people reporting that they are temporarily away from work but still

employed.

4,500,000

4,600,000

4,700,000

4,800,000

4,900,000

5,000,000

5,100,000


Fewer hours also feeds through into lower pay, although clearly falls in pay have been

cushioned by the 80% salary replacement paid through the furlough scheme.

Nonetheless, nominal pay recorded a small year-on-year fall in April, for the first time

since pay survey data was collected in 2000.

Figure 13 Average hours worked per week


Conclusions and implications

Today’s labour market data presents a somewhat confusing picture of the impacts of the

crisis over recent months.

It appears to us that the very large rises in claimant unemployment are largely reflecting

what is happening in the real economy – with HMRC data recording a fall in employee

numbers of at least 600 thousand (which may well be revised up) and the Labour Force

Survey suggesting that self-employment fell by at least 300 thousand in April. We would

expect to see this translating into rising unemployment, rather than economic inactivity, as

those out of work start looking for new jobs (as has likely already happened). Non-

claimant youth unemployment is also likely to significantly add to the numbers seeking

work. Nonetheless it is likely that the claimant unemployment rises somewhat overstate

the real level of unemployment as it was in April and May.

Looking ahead, it is certain that claimant unemployment will rise above three million next

month and may well be higher than the peak of 3.1 million reached in the 1980s

recession.

Rising unemployment has also been accompanied by a collapse in job vacancies – from

around 800 thousand before the crisis to just 320 thousand in the single-month estimate

for May. Putting this together with the unemployment data, this means that there are now

27.5

28.0

28.5

29.0

29.5

30.0

30.5

31.0

31.5

32.0

32.5

Apr-

18

May-1

8

Jun-1

8

Jul-18

Aug-1

8

Sep-1

8

Oct-

18

Nov-1

8

Dec-1

8

Jan-1

9

Feb-1

9

Mar-

19

Apr-

19

May-1

9

Jun-1

9

Jul-19

Aug-1

9

Sep-1

9

Oct-

19

Nov-1

9

Dec-1

9

Jan-2

0

Feb-2

0

Mar-

20



8.5 unemployed people competing for every job opening, compared with just 1.5 before

this crisis began. This picture also varies hugely across the country – with our analysis of

online adverts for the Joseph Rowntree Foundation suggesting that in many ex-industrial,

inner city and coastal areas the picture is far worse.

Looking ahead, new claims to Universal Credit appear to have fallen back to close to pre-

crisis levels, while data that we will publish on Friday will show that the number of new

vacancies has doubled in the last fortnight (albeit from a very low base). Both of these

are positive signs. However on the other hand unless people are confident to start

spending again, and businesses are able to reopen, then there is a very real risk that a

‘second wave’ of unemployment could hit us as the Job Retention Scheme starts to

unwind. This could push unemployment up by at least a further two million, and see it

rising inexorably towards four or even five million by the end of year. As with everything

in this crisis, all roads lead back to successfully suppressing the virus.

In the meantime though, our view is that we need to act now to deal with this jobs crisis.

As we set out over the weekend with over 200 business leaders, politicians and charities,

we think that this response needs three parts:

■ An urgent and temporary stimulus to kick-start hiring and support job retention over the

summer, ideally by cutting employer National Insurance;

■ Doubling of the size of our employment services, so that everyone gets the help that

they need to find new work; and

■ Investment now so that we are ready for record long-term unemployment next year – in

particular can guarantee that all young unemployed people will have the opportunity of

an education place, apprenticeship or job.

About IES

The Institute for Employment studies is an independent, apolitical centre of research and consultancy in employment policy and human resource management. It works with employers, government departments, agencies and professional and employee bodies to support sustained improvements in employment policy and practice.

Institute for Employment Studies, City Gate, 185 Dyke Road, Brighton, BN3 1TL United Kingdom

www.employment-studies.co.uk

@EmploymtStudies

01273 763400

https://www.employment-studies.co.uk/resource/weekly-vacancy-analysis-vacancy-trends-week-ending-24-may-2020

https://www.employment-studies.co.uk/resource/weekly-vacancy-analysis-vacancy-trends-week-ending-24-may-2020

https://www.communitiesthatwork.co.uk/opportunity-guarantee/

http://www.employment-studies.co.uk/

Documents

Labour Market Statistics, June 2020...Labour Market Statistics, May 2020: Claimant count analysis number of hours. While there does appear to be a discrepancy between the claimant