Active Labour Market Policies: Theory and Danish Experiences Torben M. Andersen Department of Economics and Business Aarhus University IZA/World Bank/OECD Conference on Activation and Employment Support Policies Istanbul 2012
Active Labour Market Policies: Theory and Danish
Experiences
Torben M. Andersen Department of Economics and Business
Aarhus University
IZA/World Bank/OECD Conference on Activation and Employment Support Policies
Istanbul 2012
Active labour market policies/workfare (ALMP/WF):
• Generous social benefits – can incentive structure be maintained by ALMP?
• More acceptance of ALMP than benefit reductions?
• Work first: Target transfers to those working, but having low income (working poor)
• Easier to get support for a ”pro-active” social safety net
Activity/participation requirements as part of the eligibility conditions for transfers (unemployment benefits, social assistance)
Aims of ALMP • Targeting: Ensuring that transfers go to the intended
recipients (“deserving”) • Job-search incentives:
Strengthen incentives to search for regular jobs (job search intensity, and reservation demands/wages)
• Qualifications:
Address skill constraints/barriers (technical, cognitive, non-cognitive/social) lowering job finding chances
Targeting – the selection problem
• Target transfers to individual with low earnings ability • Unconditional transfers: Some high-ability individuals
may claim benefits • ALMP-conditions: Higher opportunity cost for individuals
with high earnings ability to claim transfers - high-ability types self-select not to claim transfers - transfers better targeted to the ”deserving” • Possible to implement higher transfers for given fiscal
costs
ALMP and labour supply
Extensive margin: • Transfers not a
”passive” alternative to work
• ALMP lowers the
reservation wage
• Higher labour force participation for given transfer and tax levels
Labour supply
Marginal effective tax rate on work
ALMP
ALMP and targeting • Workfare does not have to be productive (direct treatment
effect) to affect labour market outcomes! • It is a screening device! • Screening paradox:
– The target group for transfers is exposed to the ordeal of ALMP – The problem was the ”non-deserving” claiming the transfer – The mechanism works by making it unattractive for the non-target
group to claim the transfer!
• Welfare case for ALMP? – Higher transfers – Disutility from ALMP (less leisure time)
Job search
• Trade-off insurance vs incentives
• Generous benefits = insurance protection income/consumption
• Reduces incentives to search for jobs (search intensity, reservation demands to jobs)
Search and ALMP • Search
– Prior to participation: more search (threat/motivation effect)
– During participation: less search (locking-in)
• Wages – outside option is changed = wage moderation
• General equilibrium effects: search and
wages affect job-creation
ALMP and employment
• ALMP improves employment for given benefit levels
• (Utilitarian) Welfare may improve
• Optimal policy – small intensity (few) but demanding activity Replacement ratio
Workfare requirement
Qualifications
• Overcome barriers for job-finding – Individual characteristics and market
developments – Targeting problem (who and what?)
• Duration dependent loss of human capital
– Strengthens the incentive to search up-front – ALMP may be a carrot (will solve my problems)
• Ex post to participation – Improved qualifications – Changed reservation demands
Activation
Threat/motivation effect: Active job search/ lower reservation demands
Screening: Deserving vs non-deserving
Locking-in Less active job search
Improved qualifications
More narrow job search
General equilibrium effects via wage formation and job-creation
Unem
ployment spell
Programmes are costly time profile: stepping up
• How to avoid dead-weight losses?
• Programme assignment (profiling)
• Adminstration – incentives and monitoring
The Danish case: Low unemployment and extensive use of ALMP
0
2
4
6
8
10
12
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
Denmark
OECD 0
0,2
0,4
0,6
0,8
1
1,2
1,4
1,6
1,8
Denm
ark
Belg
ium
Net
herla
nds
Swed
enG
erm
any
Fran
ceFi
nlan
dIre
land
Spai
nAu
stria
Port
ugal
Pola
ndLu
xem
bour
gJa
pan
Hung
ary
Italy
Kore
aCa
nada
Slov
enia
Uni
ted
King
dom
Aust
ralia
Chile
Esto
nia
Czec
h Re
publ
icSl
ovak
Rep
ublic
Uni
ted
Stat
esIs
rael
Mex
ico
ALMP costs, % of GDP, 2009
Boom-Bust pattern: Larger GDP drop than OECD average
Unemployment
The three pillars of the Flexicurity-model
• Lax employment protection= flexibility for firms
• Generous unemployment
benefit scheme = insurance/ security for workers
but reduces incentives to be
actively searching for jobs • The model presumes that
unemployment is a temporary state
• Balanced via ALMP • :
Flexible hiring/firing rules
Unemployment Insurance Active labour
market policy
Labour market reforms in the 1990s
• Shorter duration of the benefit period
• Stricter eligibility conditions
• Activation (workfare) - Condition for eligibility - Not qualifying for UIB
Short term insurance
elements maintained, but incentives strengthened
Flexible hiring/firing rules
Unemployment Insurance Active labour
market policy
Replacement rate %
Activation requirements
M o n t h s h s
6 12 18 24 30 36 42 48 54 60 66 72
90
Unemployment benefits
Social assistance
Full time activation
Activation offer: right and duty
AcActiveaastive job se Active job search
Activation offer: right and duty
Active job search
Active job search
Low income groups
APW 60
NB duration period shortened to 2 years + stricter eligibility conditions
The Danish System: Unemployment insurance benefits, Social assistance and ALMP
The Danish labour market
• High level of job-turnover
• High incidence of short term unemployment
• ”Entry friendly” Relatively low level of long-term and youth unemployment
• High level of perceived
job security (=ability to find a job)
AUT
BEL CHE
CZE DEU
DNK ESP FIN
FRA GBR
GRC
HUN
IRL
ITA NOR
POL
PRT SVK
SVN
SWE
TUR
USA
10,0
15,0
20,0
25,0
30,0
35,0
10,0 15,0 20,0 25,0 30,0
Hirin
gs %
Separation %
Job creation and destruction
High incidence of unemployment but short
duration
0
50.000
100.000
150.000
200.000
250.000
300.000
350.000
400.000
450.000
Unemployment : Affected and full-time equivalents
Number of persons affected by unemployment
Full-time equivalents
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
0
5
10
15
20
25
Average duration - left scale
Affected relative to unemployment full-time equivalents-right scale
Average duration and ratio of affected relative to full-time equivalents
Quick transition out of unemployment – also in the
current crisis
0
20
40
60
80
100
120Survival rate unemployment
2006.4-2007.3
2009.4-2010.3
360.000 spells 520.000 spells
0
10
20
30
40
50
60
3 months 6 months 9 months 12.months
%
In employment after initiated unemployment spell
2009.4 - 246.000 initiatedspells
2006.1- 160.000 initiatedspells
Low level of long-term and youth unemployment
0,0
5,0
10,0
15,0
20,0
25,0
30,0
35,0
40,0
45,0
Switz
erla
ndN
ethe
rland
sAu
stria
Japa
nN
orw
ayG
erm
any
Aust
ralia
Denm
ark
Luxe
mbo
urg
Cana
daIc
elan
dO
ECD-
Tota
lN
ew Z
eala
ndU
nite
d St
ates
Uni
ted
King
dom
Finl
and
Port
ugal
Belg
ium
Fran
ceSw
eden
Italy
Irela
ndG
reec
eSp
ain
Youth unemployment rate, % of youth labour force (15-24)
0,0
10,0
20,0
30,0
40,0
50,0
60,0
New
Zea
land
Nor
way
Cana
daSw
eden
Aust
ralia
Denm
ark
Icel
and
Finl
and
Aust
riaN
ethe
rland
sU
nite
d St
ates
Luxe
mbo
urg
OEC
D-To
tal
Uni
ted
King
dom
Switz
erla
ndJa
pan
Fran
ceG
reec
eSp
ain
Ger
man
yIta
lyBe
lgiu
mIre
land
Port
ugal
%
Long-term unemployed, % of total unemployment
ALMP – requirements and activities
• Up-front job search requirement
• ALMP within first 9 months, different rules for the young
• Activation rate = 1/3 for unemployed
• Most programmes are short (< 13 weeks)
• Most in work related ALMP (intership, employed with wage subsidy)
• Continous change in requirements and contents – Business cycle situation – Evidence
• Recently much focus on interviews and monitoring of jobs search
• Increasing demands on the young – Youth package – earlier and
more demanding ALMP (=education)
– Immediate full-time activation for the very young
ALMP -administration • Specific institutional structure: Unemployment
insurance funds (Unions) and municipalities
• One-stop shop - Job-centres (municipalities): For all unemployed
(Social Assistance and Unemployment benefits) - Easier for the unemployed • Incentive structure for job-centres - Reimbursed based on quantitative measures. What
about quality? - Effect on job counselling (area vs skills) - More lean administration?
Requirements and programme assignment: Match-groups
• Group I (job-ready): No problems except unemployment.
• Group II (ready for activity): Not ready to start
working, but is capable of participating in a program activity aiming at later employment
• Group III(temporarily passive): Neither ready for a job nor for participation in a program activity aiming at later employment
The achilles heel: Long-term unemployment
• Extended welfare state critically dependent on maintaining a high employment rate!
• Financial crisis = increasing inflow into ALMP • How to maintain programme effectiveness?
• How to identify problems (different composition
of the pool of unemployed, structural shifts)? • Maintaining political support for ALMP when
unemployment is high?
Concluding remarks • ALMP has important effects on the incentive structure in
the labour market • Can be used to balance incentives with insurance
(distributional goals) • Programmes are costly – programme administration and
assignment are critical • Danish experience: - ALMPcan be used to support a flexible labour
market which is ”entry-friendly” - Continuous adaptation of policies to curb costs and
maintain effectiveness - It has coped reasonably with the crisis – implicit
work sharing - Can it prevent a significant increase in long term
unemployment?