A multidimensional approach in the measurement of underemployment Giorgina Brown, Federica Pintaldi (ISTAT, Rome)

A multidimensional approach A multidimensional approach in the measurement of in the measurement of

underemploymentunderemployment

Giorgina Brown, Federica Pintaldi (ISTAT, Rome)

Time-related underemploymentTime-related underemployment

• core indicator within decent work framework: – employment opportunities – adequate earnings and productive work– decent hours

• within labour force framework, indicator of underutilization of labour resources (similar and complementary to unemployment)

Time-related underemploymentTime-related underemployment

• one of the 20 Key Indicators of the Labour Market (ILO)

• two rates are presented: – number of persons in time-related underemployment

1.as a percentage of the labour force

2.as a percentage of total employment

DefinitionDefinition

• “insufficient hours of work in relation to an alternative employment situation that a person is willing and available to engage in” (16th ICLS, 1998)

• Operationally, it identifies employed persons who in the reference period:– were willing to work additional hours– were available to work additional hours– had worked less than a threshold relating to working

time

Determining persons in time-related underemploymenttime-related underemployment

Employed population22.404

In reference

week willing to

work

less hours1.307

additional hours1.344

same hours19.753

Available to work more?

yes, available

1.050

not available

295

Italian LFS, 2004 (‘000s)

• Includes: – persons not immediately available because of the notice

needed to change job

• Excludes:– persons who have worked less than usual because of illness,

strike, compulsory maternity leave, or study not organized within one’s job

Determining persons in time-related underemploymenttime-related underemployment

Employed population22.404

In reference

week willing to

work

less hours1.307

additional hours1.344

same hours19.753

Available to work more?

yes, available

1.050

not available

295

Italian LFS, 2004 (‘000s)

Defining a thresholdthreshold:: – Legal or empirical– Usual or actual– Statistic used

For defining a thresholdthreshold: mean hours worked

Seasonal factors

 Variable used

QuarterYear 20041st 2nd 3rd 4th

actual with absent 38 40 34 38 37

actual without absent 40 41 41 40 41

usual 40 41 41 40 41

mixed all absent 41 42 42 41 42

mixed some reasons 41 41 41 41 41

the median has always the same value: 40 hours

Excludes some employed

Framework

Different dimensions

500

600

700

800

900

1,000

1,100

usual hours actual hours mix1 hours mix2 hours

40 hours

Time-related underemployed by thresholdthreshold and variable (‘000s)

Employed population22.404

In reference

week willing to

work

less hours1.307

additional hours1.344

same hours19.753

Available to work more?

yes, available

1.050

not available

295

Determining persons in time-related underemploymenttime-related underemployment

Actual hours

worked

in additional job53

in different job100

within current job

585

does not specify

253

Work more in what way?

up to threshold

992

above threshold

58

4.4% of employment4.1% of labour force

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

actual hours full-time employed actual hours underemployed

actual hours part-time employed hours wanted by underemployed

0

25

50

75

100

Distribution of hours workedhours worked in the reference week by employed & underemployed and hours wantedhours wanted by the underemployed

Overlap between involuntary part-timeinvoluntary part-time and time-related underemployed (‘000s)

Time-related underemployed

yes no total

Full-time 541 19.022 19.563

Involuntary part-time 358 656 1.014

Other part-time 93 1.735 1.828

Total 992 21.413 22.404

Only 36% of underemployed are also involuntary part-time

Hours worked is a more stringent and homogenous criterionconsidering the variability of working time in different periods and among occupations

• 38% of underemployed worked less than usual in reference week (vs. 17% of employed) → lack of work in specific periods.

• majority (58%) of underemployed who worked less than usual would like to work as usual, 15% state usual hours are very variable.

• But for majority of underemployed (59%), hours worked in reference week are as usual.

Always or sometimes underemployed?Always or sometimes underemployed?

1,7

1,3

3,5

1,8

3,7

1,5

1,8

2,7

3,1

2,5

1,9

2,0

3,5

2,2

Italy

services

constructions

industry

agriculture

women

men

Less than usual As usual

Always or sometimes underemployed?Always or sometimes underemployed?

Other forms ofOther forms of underemploymentunderemployment • “inadequate employment situations that affect

the capacities and well-being of workers” (16th ICLS, 1998)

• Employed persons who during the reference period wanted to change their work situation for set of reasons chosen according to national circumstances, e.g.– inadequate use and mismatch of occupational skills– inadequate income– excessive hours

»more difficult to measure »no operational definition

Skill-relatedSkill-related underemploymentunderemployment

• Our proposal of operational definition coherent with time-related underemployment:– Looking for a new job (willing)– Available to start new job (available)– Mismatch of occupation and skills (threshold)

• Level of education as proxy for skills → indications of ISCO 88:– persons with university-level degree in groups 3 to 9

(Technicians to Elementary occupations)– persons with upper secondary school diploma in

groups 4 to 9 (Clerks to Elementary occupations).

Employed population22.404

yes, a second job

216

no

20.771

yes, a new job

1.417

not available

112

yes, available

1.305

yes517

no 788

to earn more131 281

more qualifying job153 104

other reason233 433

Looking for a job?

Available to start new

job?

Mismatch of education

and occupation?

Reason for look ing for a new job

Determining persons in skill-related underemploymentskill-related underemployment

1,417

1,305

902

669621

517

284

200

400

600

800

1,000

1,200

1,400

1,600

looking for job +available +available,mismatch or

reasons (skills,earnings)

+available,reasons

+available,mismatch or

reason (skills)

+available,mismatch

+available,mismatch and

reasons

What would change if we were to use an alternative definitionalternative definition?

Mismatch only = 6,495 ‘000s

Skill-relatedSkill-related underemploymentunderemployment

• Of the 517 thousand in skill-related underemployment (2.3%):

– 413 have upper secondary school diploma(4.2% of those in employment with a diploma)

– 104 a university-level degree (2.3%).

CombinationCombination of time-related and skill-related underemployment (‘000s)

472 112 880

time-related underemployment

skill-related underemployment

Small overlap: different profiles

Proportion of underemployed by sex and ageby sex and age

0

2

4

6

8

10

12

14

all e

mploy

edm

en

women

15-3

4

35-4

4

45-6

4

65 &

ove

r

time-related both skill-related

Proportion of underemployed by type of worktype of work

0

2

4

6

8

10

12

14

16

18

20

allemployed

permanentjob

temporaryjob

self-employed

full-time part-time

time-related both skill-related

Proportion of underemployed by occupation groupoccupation group

0 2 4 6 8 10 12 14

senior off icials and managers

professionals

technicians

clerks

service and sales w orkers

agricultural, craft and related w orkers

plant/machine operators and assemblers

elementary occupations

time-related both skill-related

Other dimensions?Other dimensions?

• Income-related underemployment in part associated with time and skill. But low income possible even working many hours or for highly qualified occupations.

• To be coherent, we need variable on earned income to establish threshold of adequate income above which one cannot be classified as income-related underemployed– Measurement problems: interviewees often reticent

• To avoid overlap income-related/time-related dimensions, we could consider hourly income.

• We could distinguish those who would like to work more hours but only if paid more than their normal hourly rate, from those who want to work more even at same rate.