Roma poverty and
deprivation: the need
for multidimensional
measures
Andrey Ivanov, FRA
1
Outline
• Why monitoring progress on Roma inclusion is
important?
• What are the myths around data and what are
the available data sources?
• What the available data say – and what they do
not say?
• How multidimensional poverty monitoring might
help?
2
The EU: ‘cascading’ involvement in
Roma integration (1)
• The European Commission Communications
2010-2013
– Framing the issue in line with the Inclusive Growth
priority of the EU 2020 strategy; calling for National
Roma Integration Strategies and clear local-level
focus with active role of Roma civil society; assessing
the first drafts of the strategies and their results.
• The European Council
– 2013 Council Recommendation on effective Roma
integration measures in the Member States and on
monitoring and evaluation
3
The EU: ‘cascading’ involvement in
Roma integration (2)
European Parliament and the Council
• 2013: Ring-fence allocation of 20 % of the total
ESF resources for “promoting social inclusion,
combating poverty and any discrimination” and
• Ex-ante conditionalities for improvement of the
situation of marginalised communities such as
the Roma
– the implementation of a National Strategic Policy
Framework for Poverty Reduction (Conditionality 9.1)
– a national Roma Inclusion Strategic Policy Framework
(Conditionality 9.2).
Three myths about data on Roma
1. There is no data, so we don’t know
•
A number of countries use ethnic markers on censuses
and standardized European social surveys
Territorial mapping
Custom surveys
•
•
2. There is no need of data because we know how
bad it is anyway
•
It is important to know not just how bad it is – but most of
all, why?
3. We might need but it can’t be collected because of
legal constraints
•
Constraints exist but they are overestimated
Measuring progress
• Progress of who?
– Defining the target group is misleadingly
– The outcome differs depending on the approach one
takes
• Research (historical or ethnological)
• Pragmatic (policy-driven)
• Progress in what – integration vs. inclusion
• Measuring how? What indicators to populate with
the data?
– Input-output-outcome
– Structure-process-outcome
6
Data examples: census, Bulgaria (1)
Gross enrolment rate in primary, secondary and tertiary educational levels
(share of the respective age group)
7
Source: NSI, census 2011
Data examples: census, Bulgaria (2)
Highest achieved educational level (population
aged 7 and above who are not in education)
8
Source: NSI, census 2011
Data examples: census, Bulgaria (3)
Employment, unemployment and activity rates
9
Source: NSI, census 2011
Data examples: LFS, Hungary
Labour market participation of Roma and NonRoma, 2013
10
Source: LFS, 2013
Data examples: LFS, Hungary (2)
Security of employment among Roma and NonRoma, 2013
11
Source: LFS, 2013
Data examples: custom surveys
(FRA, UNDP/WB/EC)
• Data derived from representative surveys in 11
EU Member States (FR, ES, PT, IT, PL, EL,CZ,
SK, BG, RO, HU)
• Two samples
– Roma
– Their non-Roma neighbours
• Levels of comparability:
– within groups,
– between groups,
– with national averages (on major indicators)
12
Population at risk of poverty
Population in household with equivalent expenditure below 60% of the national median, in %)
Sources: FRA Roma survey 2011, EUROSTAT 2011
Structure of household income, 2011
Roma
Non-Roma
The share of total work-related incomes (from “employment” and “other labour related
activity”) is remarkably similar between both groups. The same applies for social transfers
with the only difference in pensions (higher share among non-Roma) and social assistance
(higher share among non-Roma)
Source: UNDP/WB/EC Roma survey 2011
Structure of household expenditures, 2011
Roma
Non-Roma
The income structure of Roma households is dominated by expenditure on
food – typical for developing countries
Source: UNDP/WB/EC Roma survey 2011
Attainment rates
25 to 64 who completed at least upper secondary education (vocational or general) (%)
Sources: FRA Roma survey 2011 - LFS 2011
School (un)attendance
Respondents aged 16 and above who have never been to school (%)
…or in other words – who have
not had the chance of
exercising their fundamental
right to education
Source: FRA Roma survey 2011
Malnutrition
Share of persons living in households in which someone went hungry at least once last
month because the family couldn’t afford buying food
High share of
Roma families
cannot exercise
a fundamental
right of being
free from
hunger
Source: FRA Roma survey 2011
Unemployment rates by gender
and ethnicity
19
Source: UNDP/WB/EC Roma survey 2011, EUROSTAT
Discrimination and prejudice
Experience of discrimination in employment in the
last five years because of being Roma in the 5 EU
Member States (%)
20
Source: FRA Roma survey 2011
We have the data…
Now what?
Having the data is just the first step. Making sense of it (and using it for policy
purposes) does not come automatically
21
Roma Multidimensional poverty index
• Follows Alkire and Foster (2007) methodology
• Structured in two areas
a. Human capabilities
b. Material wellbeing
• Six equally weighted dimensions
a.1. Basic rights
a.2. Health
a.3. Education
b.1. Housing
b.2. Standard of living
b.3. Employment
• 12 indicators (two for each dimension)
22
‘Human capabilities’ area
Dimension
Criterion of deprivation and threshold
Level of
observ.
Civil status
Having an ID – yes/no (personal document, birth
certificate etc.)
I
Discrimination
HH member lives in a HH where a member has been
discriminated against while looking for a job
P
Disability status
A household member having a disability – yes/no
I
Indicators
Basic rights
Health
Education
Any HH member living in a HH responding "yes" to the
Limited access to medical question "were there any periods in the past 12
months when you couldn’t visit a doctor when you
services
needed?”
For adults: any HH member above schooling age who
hasn’t completed primary education or lower
Highest
completed secondary
education
For children: children in school age who are not in
school
P
I
23
Self-declared illiteracy rate Any HH member stated as unable to read and write
I
‘Material wellbeing’ area
Dimension
Indicators
Access
to
infrastructure
Housing
Level of
observ.
A composite indicator –any HH member living in a HH
without two of the three (toilet or bathroom inside the
house; running water; electricity)
H
Shares
of
the
population not having
Any HH member living in "ruined houses" or "slums"
access
to
secure
housing
H
Any HH member living in a HH that experienced that in
the past month somebody ever went to bed hungry
because they could not afford enough food for them
H
Any HH member living in a HH, which doesn't possess
Access to various HH
four of six categories falling in the "Material
amenities
deprivation" index
I
Extreme poverty
Standard
living
basic
Criterion of deprivation and threshold
of
Unemployment
Any HH member living in a household with none of the
adult HH members employed (16+).
H
24
Measuring poverty – but which exactly?
Source: UNDP/WB/EC Roma survey 2011
26
27
28
Composition of Roma poverty (BG, RO)
Non-severe and severe poverty rates bars left
scale) and the value of MPI (right scale)
29
Source: UNDP/WB/EC Roma survey 2011
Multidimensional poverty 2004-11 (BG, RO)
Change in multidimensional poverty rates of Roma and non-Roma
30
Sources: UNDP Roma survey 2004; UNDP/WB/EC Roma survey 2011
Structure of deprivations
Changes in multidimensional poverty deprivations
structure of Roma in BG and RO, 2004-2011
31
Source: UNDP/WB/EC Roma survey 2011
Quantitative data is not enough
• Data outlines the status and rarely the
determinants
• Figures are rarely put in their specific context
• We keep measuring what is measurable
• The missing dimensions
– Agency
– Fundamental rights
– Discrimination
32
The missing dimensions: aspirations
Educational aspirations and multidimensional
poverty
33
Source: UNDP/WB/EC Roma survey 2011
Conclusions - data
• Data from different sources have different strengths
and weaknesses and should be used in
complementary manner
• Including ethnic identifiers makes possible
– For censuses to provide reliable and robust data for
monitoring long-term changes
– For standardized European surveys to yield data with
higher frequency
• Custom sample surveys can
– Fill the gap in cases when applying ethnic identifiers is not
possible and
– Provide comparability across countries
34
Conclusions - indicators
• Multidimensional poverty concept reflects better
the specific challenges of Roma inclusion
• It yields lower poverty rates but reflects the
reality better
• From policy perspective, it allows understanding
better the drivers of poverty
• Important dimensions (namely agency and
aspirations) are still not sufficiently covered
– An area that might be addressed through thematic
modules in the standardized European Surveys
35
Thank you for your attention!
For more information you can
contact us at
romaprogramme@fra.europa.eu