Activation and Smart Safety Nets
in the Western Balkans
Boryana Gotcheva & Aylin Isik-Dikmelik
Vienna, March 4, 2014
Outline
 Why activation? What is the challenge?
 What is activation?
 The Western Balkans Activation and Smart Safety Nets
Study:
 Content
 Outputs
 Analytical framework
 Going forward
The Challenge
 Employment and active inclusion are among the most
critical challenges for countries across the Western
Balkans
Kazakhstan
Netherlands
Azerbaijan*
Austria
Denmark
UK
Germany
Luxembourg
Czech Republic
Estonia
Georgia*
Slovenia
Portugal
Albania*
Ireland
Latvia
Slovakia
France
Poland
Lithuania
Belarus*
Romania
Belgium
Hungary
Bulgaria
Spain
Turkey
Italy
Greece
Armenia*
Montenegro
Croatia
Moldova
Serbia*
FYR Macedonia
B&H
Kosovo
Percentage
Daunting jobs challenge with low activity and
employment rates …
80
Employment (age 15+), Q4, 2011
70
60
50
40
30
20
10
0
…and unemployment rates that are among the
highest in Europe
Unemployment rate (latest available)
25
20
15
10
5
0
Unemployment rates, 2008-2012
CIS
Western Balkans
Non-ECA Europe
Baltics
ECA
What is activation?
 Activation has a ‘history’ across the world
 Shift from “passive” to “active” social policies
 Primary focus on employability and labor market re-attachment
 The “European” model of activation policies generally
place employment integration at the very heart of social
policy
This implies:
 Greater emphasis on work as a way to ensure that individuals in
their prime age are not excluded from mainstream society
 Greater effort by the social and employment services in helping
inactive, unemployed and benefit recipients overcome the
obstacles to entering into paid work
 Greater effort by the inactive, unemployed, and recipients to
enter or re-enter the job market
Key building blocks of activation
1. The ‘mutual obligations’ principle
2. Frequent and personalized/tailored interventions of agencies
during individual’s unemployment spell
3. Financial incentives to labor market re-integration
4. Stricter benefit eligibility criteria and benefit sanctions
5. Strong obligations to participate in labor market programs,
community work or, to try self-employment
6. Changes in institutional arrangements and greater
coordination across institutions
Promoting Employability and reducing reliance on
social transfers
Activation “Package” of incentives & support services
Integrated
Service Models
Tailored to
Clients
Employment &
Benefit Incentives
Social Service
Supports
Activation is for both the inactive and those who
are active but without a job
WITHOUT A
JOB
INACTIVE
Out of the LF
but active
Non-Poor
Poor
SSN-Beneficiaries
Increase
Productivity
WITH A JOB
The Western Balkans Activation and Smart
Safety Nets Study
Builds on
Country-specific
previous
case studies for
Advice to the
Assistance on
research Social
all Western
Governments on
how to
Protection
Balkans
policy changes
implement
Systems in
including entity-
that promote
activation
Western
level cases in
activation
policies
Balkans
BH
Framework for the analysis
• Target Groups?
• Inactive
• Unemployed
• SSN Beneficiaries
• Barriers to Work?
• Employability
barriers (skills,
experience, etc.)
• Participation
constraints
Activation for Who?
PROFILING
(Dis)Incentives in
Benefit Design
• Benefit formula
• Generosity/’package’
• Mutual obligations
• Duration/ phase out
• (Dis)incentives in tax
• Coordination between
welfare and
employment services
• Specific activation
policies and ALMPs
• Implementation
capacity (financing,
staffing, etc.)
and benefit systems
• Earned income
disregards
Institutional Readiness
for Activation Policies
Coverage of countries and activities
Country
Profiling
(Dis)incentives in
the design
Phase 1-2 (2012-13)
FYR Macedonia
Kosovo
Montenegro
Serbia
Phase 3 (2013-14)
Albania
Bosnia and
Herzegovina
Institutional
capacity
Activation for Who? Profiling: SERBIA
• Target Groups?
• Inactive
(Dis)Incentives in
Benefit Design
• Unemployed
• SSN Beneficiaries
• Barriers to Work?
• Employability
barriers (skills,
experience, etc.)
• Participation
constraints
Activation for Who?
Institutional Readiness
PROFILING
for Activation Policies
Analytical framework to analyze constraints to
employment of safety nets beneficiaries
Employability
barriers
Tax and Benefit
Disincentives
Participation
constraints
These barriers are interrelated
Objective of “Profiling” of Social Safety Net
beneficiaries
 Providing tailored activation strategies for a
diverse vulnerable population
Who can be activated in the population?
Who are the “activable” among SSN beneficiaries?
Profile of the “activable”
What are the employability barriers?
Are there additional constraints to their participation ?
What are the main client segments for activation?
What graduation strategies suit each group?
Who can be “activated”?
Of working age (15-64)
Able bodied
Activables:
Individuals who can be presumed
to be able to work
Not in education or training
• Who can be activated among the population?
• Who can be activated among the SSN beneficiaries?
• Are these groups coinciding?
More than half of population in Serbia are “work-able”
(potentially “activable”)
Age Composition of SSN Beneficiaries Relative to General
Population in Serbia, 2010
FSA beneficiary
26.4
FSA+CA
beneficiaries
11.0
11.3
SSN all
32.1
27.1
Whole population
12.6
0
4.7
54.8
6.1
9.6
22.1
20
50.0
8.6
52.9
8.0
56.8
40
60
80
100
Percent
Child
Old
Working age (disabled)
Working age (in education)
Source: Serbia HBS data 2010.
Note: “Work-able” includes all individuals of working age (15–64) who are neither disabled nor in education or training.
Working age (work-able)
SSN beneficiaries represent only a small fraction of
the work-able population
Focusing only on SSN beneficiaries will have limited impact in the LM
Safety Net Coverage of the Work-Able Population in Serbia, 2010
Out
of labor force
72.8
Unemployed
13.5
9
83.3
Employed
8.5
79.0
0
20
13.0
40
60
80
5.
6
1.9
6
100
Percent
Nonbeneficiaries in Q2-Q5
Nonbeneficiaries in Q1
Beneficiaries of other SSN
Source: Serbia HBS data 2010.
Note: “Work-able” includes all individuals of working age (15–64) who are neither disabled nor in education or training.
FSA beneficiaries
SSN Beneficiaries more likely to be unemployed or
inactive or have low-quality jobs
Employment and Unemployment rates among
the work-able population in Serbia, 2010
FSA
beneficiaries
56.5
FSA + CA
beneficiaries
58.1
SSN beneficiaries
20.7
Nonbeneficiaries, poor
22.8
21.9
56.8
FSA
beneficiaries
56.1
20.1
20.7
55.2
Sector of Employment for work-able
Population in Serbia, 2010
SSN beneficiaries
13
Nonbeneficiaries
, poor
17.2
25.9
63.0
27.8
20
16.4
40
60
Unemployed
14.7
40.3
34.4
28.6
11.2
25.7
20.6
80
Percent
Employed
44.1
18.8
0
0
12.7
22.4
Whole
population
Whole population
30.3
Out of labor force
100
20
40
60
Percent
80
Public and professions
Retail, trade, crafts
Constr., industry, transport
Agriculture and manual jobs
Not identified*
Source: Serbia HBS data 2010.
Note: “Work-able” includes all individuals of working age (15–64) who are neither disabled nor in education or training
* Because of the sample size, conclusions cannot be drawn about the sectors other than “Agriculture and manual jobs.” .
100
Which could be largely explained by lower educational
attainment
Education Distribution of SSN Beneficiaries in Serbia, 2010
FSA beneficiaries
17.3
SA + CA beneficiaries
33.5
12.1
Nonbeneficiaries,
poor
28.5
9.0
SSN beneficiaries
5.4
0
52.5
29.2
11.0
Activable
population
40.8
56.4
28.0
53.0
19.0
10
59.3
20
30
40
Never attended
50
Percent
No education completed
Secondary/Vocational
Higher education (college or higher)
60
70
80
Elementary school
Source: Serbia HBS data 2010.
Note: “Work-able” includes all individuals of working age (15–64) who are neither disabled nor in education or training.
90
100
Work-able SSN beneficiaries display greater caretaking
needs than the work-ready population as a whole
Share of work-able population living with at least one person in need of care
in Serbia, 2010
15.4
FSA beneficiaries
29.8
23.7
2.4
FSA+CA beneficiaries
8.2
4.9
9.5
SSN all
31.5
20.4
2.6
General population
9.4
5.3
0
5
10
15
20
25
30
Percent
% hh with disabled
% hh with child ≤5
% hh with child ≤2
Source: Serbia HBS data 2010.
Note: “Work-able” includes all individuals of working age (15–64) who are neither disabled nor in education or training.
35
Latent Class Analysis: SERBIA
Inactive
uneducated
women, 21%
Elder experienced
unemployed, 35%
Elder experienced
inactive, 16%
Educated
unemployed
youth, 8%
Chronic
unemployed, 8%
Inexperienced
unemployed
women,, 12%
These groups display different employability &
labor supply constraints…
(high)
Employability
Obstacles
(Skills,
Experience)
Other Barriers to Participation
(Social, Circumstances, Other)
(high)
23
… Which call for different packages of
services
Size= % of total work-able FSA beneficiaries
4
3.5
Employability obstacles
Hard-to-serve
(skills, special support)
Intensified Activation
(TVET, Skills)
Chronic
unemployed
3
Elder
experienced
unemployed
2.5
Inactive
uneducated
women
Inexperienced
unemployed
women
2
1.5
1
0.5
Educated
unemployed
youth
Experienced
inactive elder
Special Support
(care for dependents,
transport, social, health)
Market Ready
(job info, matching, search
assistance)
0
0
0.5
1
1.5
2
2.5
Other barriers to participation
3
3.5
(Dis)Incentives in Benefit Design
(Dis)Incentives in
Benefit Design
• Benefit formula
• Generosity/’package’
• Mutual obligations
• Duration/ phase out
• (Dis)incentives in tax
and benefit systems
• Earned income
Activation for Who?
PROFILING
disregards
Institutional Readiness
for Activation Policies
FYR Macedonia Social Financial Assistance
Type of program
• Minimum income
Design, financing and
implementation
• Centrally designed: by the
Basic administrative data
• Number of beneficiaries:
guarantee / last resort
Ministry of Labor and
35,450 (2012) beneficiary
social assistance
Social Policy
families , encompassing
• Granted to individuals (and
their households) that are
• Centrally financed
• Implementation at local
115,000 direct and indirect
beneficiaries
able to work but cannot
level: by inter-municipal
provide for themselves
Social Work Centers which
(about 23 million USD)
materially
are de-concentrated
(2012)
• Works as a substitute for
structures of MLSP
• Spending: about 1.1 billion
• Spending is around 0.3
the contributory
percent of GDP (2011),
unemployment benefit
which is at the regional
which is received by only 9
average, but declining
percent of the unemployed
since 2007
SFA design implies both incentives and
disincentives to work
Incentives
Disincentives
Work incentives exist mainly as work
requirements
Job search, participation in ALMPs and training, job
interviews; job offers; municipal PWP and seasonal jobs
Monitoring of compliance is relatively strict
There are sanctions for non-compliance with work
requirements, but the right to FSA can be transferred
Additional incentives exist
• SFA benefit declines with time to
• SFA beneficiaries can be
50 percent of the initial amount
engaged in public works up to
after three consecutive years of
five days per month without
receipt
losing the right of SFA
• PWP income not disregarded,
but reentry to the SFA program
easier
Declining Schedule
for Receipt of SFA
Legal Guarantee for
SFA status while on
PWP
Disincentives for work stem from the SFA benefit
formula
 The benefit is calculated as ‘difference’ between the SFA threshold
applicable to a family of that size and its monthly income
 Each additional denar earned is taken away from the benefit
amount due -> 100 percent marginal effective tax rate
 Generosity is less of an issue: core SFA benefit is not very generous
 Generosity is measured as percentage of the consumption in the
poorest quintile ‘covered/paid for’ with the benefit amount
 …but there’s ‘packaging’ with other benefits such as electricity
subsidy
 Health insurance is “de-linked” from SFA status (no longer part of
the package). Poor, including SFA beneficiaries, are eligible for
free health care, but it is provided outside the social assistance
system
Disincentives could emerge from the design of
activation measures
Participation in activation could undermine eligibility for SFA
No legal guarantees for ‘restoring’ SFA status after finishing participation in
activation, exception are PWPs
Standard re-certification rules apply, no flexibility of ‘in and out’
Incomes from training, other ALMPs, and from public works are fully
calculated in the family income
Will impact the outcomes from activation
Overall, modern social assistance programs for work
able poor have more incentives for activation
Guaranteed minimum income / last resort programs to be open for
poor of working age and able to work when they are short of income
Some incomes from work could be disregarded
Income thresholds for program exit could be higher than the entry
thresholds
Some benefits could phase out gradually / be carried over for some
time after the beneficiary gets a job (in-work)
Would help avoid the disincentive effect of high taxation of low incomes from
work
Institutional Readiness for Activation Policies
(Dis)Incentives in
Benefit Design
• Coordination between
welfare and
employment services
• Specific activation
policies and ALMPs
• Implementation
capacity (financing,
staffing, etc.)
Activation for Who?
Institutional Readiness
PROFILING
for Activation Policies
Two main service delivery mechanisms in the region…
… which are not coordinated
 Limited interaction between employment services and
centers for social work
 Cooperation varies across localities –
 better where local offices in the same or close
premises
 Lack of formal referral procedure for social assistance
beneficiaries to employment services
 High and uneven caseloads in many centers for social
work and employment offices  little room for casework
Labor market institutions in place, but capacity for
activation on a large scale still insufficient
 Limited capacity for individualized interventions
 Due to the high client to staff ratio
 Financing constraints for active measures
 Low level of spending on active policies ranging only
0.1-0.2% GDP – significantly lower than the EU-27
average
 Lack of outsourcing of the job placement and/or
counseling services to non-state providers
 The regulatory framework for outsourcing is not fully
developed
Going forward: activation agenda much broader than
just focusing on addressing welfare dependency
 Activation agenda goes beyond safety net beneficiaries:
They are only a fraction of the inactive, and activation
measures that only target them will not bring significant
impact
 Smarter design of last resort social assistance programs
needed–Enable/promote access to working poor while
building in incentives for work (e.g. introduction of gradual
income disregard, in-work benefits etc.)
Next steps in the broader activation agenda
 Substantial amount of diagnostic work in place to inform
legislative changes for future reforms
 Room for country-specific follow up work to address needs
for additional knowledge, information and provide just-intime technical assistance
 Investments to reform and strengthen the institutional set
up for the provision of differentiated and at the same time
integrated services that would help reduce multiple barriers
to work and activate broader groups of inactive
Next steps in the broader activation agenda
Possible institutional reforms for promotion of activation
• Improving the capacity for planning and designing activation
measures, and evaluation of their effectiveness
• Improving the capacity and effectiveness of the public
employment services for implementing activation measures
• Strengthening inter-institutional cooperation – especially
between the employment services and the centers for social work
for a holistic approach to effective activation of vulnerable
• Improving the cost-effectiveness of the ALMPs and other
interventions – e.g. increased competition, advanced (statistical)
profiling, application of job matching software tools, etc.
THANK YOU!
Boryana Gotcheva bgotcheva@worldbank.org
Aylin Isik-Dikmelik aisikdikmelik@worldbank.org
Latent Class Analysis: SERBIA
Elder experienced
unemployed
Statistics
Active covariates
Indicators
Class size
Inactive
uneducated
women
Elder
experienced
inactive
Inexperienced
Educated
unemployed Chronic
unemployed
women
unemployed youth
35%
21%
16%
12%
8%
8%
Worked before
100%
19%
95%
24%
20%
21%
Willing to retrain
54%
23%
5%
45%
66%
73%
Inactive
22%
100%
100%
16%
0%
19%
Long-term unemployed
66%
0%
0%
63%
99%
6%
Short-term unemployed
Uneducated
Elementary education
12%
6%
34%
0%
31%
54%
0%
21%
36%
21%
31%
38%
0%
20%
37%
75%
6%
21%
Secondary+ education
61%
16%
43%
31%
43%
73%
Young (15–29)
4%
41%
8%
26%
39%
92%
Adult (30–54)
Prime age (55–64)
Female
54%
42%
41%
52%
7%
82%
45%
47%
34%
59%
15%
92%
61%
0%
28%
4%
4%
26%
Caretaker
Married
Discouraged inactive
(% of total)
0%
62%
20%
33%
65%
56%
32%
64%
78%
0%
48%
8%
0%
55%
0%
0%
9%
14%
Willing inactive
(% of total)
Mean age
2%
44%
22%
8%
0%
5%
46
32
47
36
31
23