Activation at work in Belgium Employment pathways out of social assistance

Activation at work in Belgium Employment pathways out of social assistance

Sarah CarpentierHerman Deleeck Centre for Social Policy Antwerp University

MIPI meeting Stockholm, 15-16 June 2010

2

1. Introduction 2. Activation in the Belgian social assistance scheme3. Data 4. Methodology 5. Hypotheses6. Preliminary results 7. Tentative conclusions8. Further Analysis

Outline

3

1. IntroductionResearch questions • Which employment pathways can we discern in the exit

from social assistance to paid labour? - Regular labour market versus labour market

integration programmes • Who has which employment pathway

- Individual socio-economic characteristics- Labour market history - Duration in social assistance

• Which are the characteristics of the employment along different pathways- Full time/ part time employment - Full/partial independency from social assistance

4

1. IntroductionLiterature• Two strands of activation literature

- International comparative studies focussing on policy design(e.g. Lodemel & Trickey, 2000; Barbier & Ludwig-Meyerhofer, 2004; Serrano Pascual & Magnusson, 2007; Eichhorst & Konle-Seidl, 2008; Immervol, 2009)

- Assessment of specific active labour market programmes (e.g. Heckman et al. 1999; Martin & Grubb, 2001; Card et al., 1999)

• Literature on social assistance dynamics (e.g. Bane & Ellwood, 1986; Blank, 1989; Leisering & Leibfried, 2001; Saraceno, 2001; Capellari & Jenkins, 2008; Dahl & Lorentzen, 2003; Hansen, 2008)

5

1. IntroductionApproach• Combination of longitudinal perspective &

policy design approach by studying trajectories to multiple exit statesby survival analysis

(self) selection of beneficiaries in employment paths

6

1. IntroductionBelgium is an instructive case because of • High expenditure on active labour market

programmes • Very rich administrative data• Nation-specific in-depth analysis• Federal country

division of competences strongly impacts on social assistance

2. The Belgian social assistance scheme

8

2.1. Overall characteristics

• Last financial safety-net for able-bodied persons at working age

• In addition to - Other categorical social assistance schemes- Unemployment benefits unlimited in time - Flat-rate benefit for school leavers

• Family-based means-test• Shared competence of the federal state and

589 local welfare agencies• Indirect multilevel governance

9

2.2.Activation

2002 Social Integration Act• Aim: maximum societal integration, participation and

self-sufficiencymainly through employment

• Social assistance can be provided as - A cash benefit, in combination with a personal social

integration contract- A job in an labour market integration programme

• Working in an activation programme: =working according to labour lawgoes with loss of social advantages linked to beneficiary statute, except when entitled to supplementary benefit

10

2.2. Activation

• Availability for work, except when exempted• Instruments: incentives and services

- Generosity: rather low to international standards, especially for singles & couples with children

- Strict time limit for youngsters: 3 months - Personal integration contract

obligatory for youngstersoptional for those aged 25 and more

11

2.2. Activation

- Fixed rate socio-professional exemption in calculation of the benefit to stimulate to take up a (part time) job

- Inactivity and poverty gaps Full time: Substantial gains, but differentiated along household types (less attractive for couples with a single wage earner and lone parents)Part time: less attractive, especially for couples with a single wage earner

12

2.3. Active labour market policy

• High expenditure on active labour market policy • Focus on job creation

low expenditure for vocational training & stimuli for self-employment

13

2.3. Active labour market policy Labour market integration programmes for social assistance beneficiaries • Direct job creation programme (Article 60)

- Specific programme for social assistance beneficiaries - Welfare agency is the employer- “User” is the welfare agency itself, an non-profit organisation

or an acknowledged social economy initiative- During the number of days the beneficiary lacks to be entitled

to full social rights - Total exemption of employer’s contribution - Federal subsidy (circa 900 €/month if full time)- Minimum part time (max. 6 months); mainly full time (or 4/5)- Permanent/fixed-term contract (with clause)

14

2.3. Active labour market policy • Other labour market integration programmes for beneficiaries

common features- Accessible on the basis of proved former periods of inactivity

(for age groups)- Fixed-term exemption from employer’s contribution & federal

subsidy- Min. 1/3 or part time work - Contract: permanent/ fixed-term/ temporary3 types- Activa (all beneficiaries; all types of employers)- SINE (low-skilled; acknowledged employer in the social

economy)- Work experience (on-the-job training; regional design

component)

15

2.3. Active labour market policy Distance from labour market

Article 60SINE ACTIVAWork Experience

• Other subsidised employment programmes (no fixed-term)Service VouchersPermanent subsidised workers in local governments

16

2.4. Labour law • High employment protection for insiders

distinction between blue collar workers and employees• 3 types of contracts

- Permanent - Fixed term - For a specific task

17

3. Data • 28590 persons aged 18-64= stock population June 2005,

not working and not entitled to other benefits • 1/3 proportionally stratified sample• Administrative data from Datawarehouse

Labour Market and Social Protection • Result of the trade-off between

information on the duration in social assistance and the number of events observed

Lexus chart for stock population June 2005

18

23

28

33

38

43

48

53

58

63

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

a

g

h ij

bc

k

ef

d

19

3. Data • Individual characteristics of the sample (June 2005)

- 53% of women- 27% under 24 years old, 24% between 25-34 years- 65% has another nationality at birth; in 2005 17% of

these persons has acquired the Belgian nationality- 30% one person household, 23% lone parents,

14% couple with children, 26% other - Work intensity since 1999

72% has not worked in Belgium

Duration in social assistance

17

12

10

1018

33 1-6 months7-12 months13-18 months19-24 months 25-32 months33 months or more

21

4. Methodology

Classification of employment pathways • Compromise between 4 considerations

- What is theoretically interesting • Wage earner versus self-employed• Normal contract (FT/PT) versus temporary contract• Active labour market policy vs. regular job circuit

- Linked to (social assistance) beneficiary status- Fixed-term versus no time limit

- Which employment paths involve different processes & actors

- Possible to distinguish in the data- Substantial number of transitions

22

4. Methodology

• Five exit states- Article 60- Fixed-term active labour market programmes- Other subsidised employment/ regular labour

market - Temporary or seasonal work - Self-employment

23

4. Methodology

• Survival analysis - Late entry design

modelling calendar time, duration as a covariate- Competing risks analysis of employment pathways

(not yet integrated) - Event: exit to first employment

(can be combined with supplementary benefit)- Discrete-time framework

Complementary log log link function

24

5. Hypotheses

• Transition to work - Households with children have a lower probability to exit than

lone persons (financial incentives)- Men have a higher probability to find a job - Non-linear relationship for age

• Employment paths (active labour market programmes design)- Higher work intensity, higher probability to be employed in

the regular labour market; followed by activation programmes, and finally the article 60 programme

- Longer duration in social assistance, lower probability to exit to regular labour market, higher probability to exit to article 60 and activation programmes

6. Preliminary results

26

Survivor function

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

26 27 28 29 30 31 32 33 34 35 36

Survivor function (all employment paths) for the June 2005 stock population

Survivor functions by employment pathway for the June 2005 stock population

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

1 2 3 4 5 6 7 8 9 10 11

Article 60 programmeFixed-term activation programmeSelf-employmentTemporary work Subsidised/regular work

Temporary or seasonal work 10%

Regular labour market/ No fixed-term ALMP 14%

Article 609%

Fixed-term activation Programme3%

Social assistance beneficiaries (100% of the stock population) Self-employed

1%

Death0.6%

Pension (no supplementary social assistance benefit)0.3%

Flat-rate unemployment benefit (no supplementary SA benefit)0.2%

Other Unknown0.2%

No exit 61%

Invalidity 0%

38%

Employment pathways at first moment of employment by December 2007

25

7

39

27

3 Article 60

ALMP forbeneficiaries

Other subsidisedemployment /regularjob circuitTemporary orseasonal work

Self-employed

30

6. Preliminary ResultsControlling for other characteristics• Duration

- longer duration, lower probability of exit to article 60 and interim

- non-linear relationship of duration for activationprogrammes

- No significant effect on exit to regular LBM or self-employment

• Work intensity- higher work intensity, significant less probability to exit

to article 60- no significant effect of the work history on the

probability to exit to an activation programme- Higher work intensity, higher probability to exit to

interim/self-employment/regular or subsidised work

31

6. Preliminary Results• Household type (ref=lone person)

- Article 60: lone parent higher probability; couple with/without children & other lower probability

- Activation Programme: no significant effect- Self-employed: couples with children, higher exit probability- Interim: unknown householdtype, higher probability- Regular/subsidised: couples with children & other higher probability

• Agefor all employment paths: non-linear effect

• Sexno significant effect for Article 60 & Activation Programmeman: higher probability of exit to interim/regular work & self-employment

• Nationality at birthArt.60: non-EU lower probabilityInterim: non-EU higher probability

• Similar for type of legislation, also higher probability to exit to regular/subsidised work

Percentage entitled to supplementary benefit at first moment of work by exit state

Article 60 0

Activation programme for beneficiaries 12

Other subsidised employment/regular work 26

Self-employed 32

Temporary/seasonal work 33

Total 30

Percentage of working time at first moment of employment by employment path

0%10%20%30%40%50%60%70%80%90%

100%

Article 60 ALMP beneficiaries Temporary work Subsidisedemployment/regular

circuit

1-10% 11-30% 31-50% 51-70% 71-90% 91%-100% (and more)

34

7. Tentative conclusions• Social assistance population has a substantial

distance from the labour market • Remarkable high percentage (30%) combines

work with a supplementary social assistance benefit

• Studying all employment paths is instructive and crucial to understand a welfare-to-work scheme- Employment by active labour market programmes

accounts for only 1/3th of all exits to paid labour - High share of seasonal or temporary workers

35

7. Tentative conclusions• Employment paths mediated by individual

characteristics, in addition to work and social assistance history

• Employment paths are linked with characteristics of employment

• Conceptualising activationemployment paths useful! Labour law

36

8. Further research

• Refining the classification of employment pathways

• Survival analysis with multiple events in discrete time

• Studying the sustainability of employment for the different employment paths

37

Thank you for your attention!

[email protected]

38

Questions for discussion• Classification of employment pathways • Conceptualisation of cumulated statutes

(duration in social assistance; transition to employment)

Prob

abili

ty o

f em

ploy

men

t

Prob

abili

ty o

f em

ploy

men

t (w

ithou

t SA

)

Art

60

Fixe

d-tim

e A

LMP

Self-

empl

oyed

Seas

onal

/Tem

pora

ry w

ork

Reg

ular

labo

ur m

arke

t/oth

er A

LMP

SexMan 43 68 23 7 4 36 33Women 32 72 30 7 2 16 44Unknown 99 35 10 2 1 55 32Age at quarter of entry18-24 50 70 17 4 2 30 4825-34 47 66 21 7 3 29 3635-44 39 72 35 10 3 22 3045-54 22 74 46 8 5 19 2755-64 6 69 32 8 13 18 31WI0% 31 70 26 7 3 26 381-10% 53 65 24 6 2 30 3811-25% 56 68 25 8 3 29 3726-50% 57 76 26 7 3 24 4151-75% 56 81 20 6 6 23 4576-100% 49 83 35 9 14 11 32