ESPON TiPSE
The Territorial Dimension of Poverty and Social Exclusion in Europe
ESPON Seminar, Rome, December 4th 2014
Workshop 2.B
Regional Resilience and Poverty
Professor Mark Shucksmith, Newcastle University UK.
Introduction
The TIPSE project is the first comprehensive and systematic
attempt to map (NUTS3) regional patterns of Poverty and
Social Exclusion across Europe to inform policy decisions.
The research team comprises 7 partners from 5 EU Member States:
•
•
•
•
•
•
Nordregio, Stockholm (Petri Kahila, project coordinator)
James Hutton Institute, Aberdeen (Andrew Copus, project leader)
Newcastle University, Newcastle upon Tyne
Institute of Economics, Hungarian Academy of Sciences, Budapest
ILS – Research Institute for Regional and Urban Devt, Dortmund
EKKE – National Centre for Social Research, Athens
•
UHI Millenium Institute, Inverness
ARoP – At Risk of Poverty Rate (Eurostat)
Source: Eurostat Regio database
Table ilc_li41
Reykjavik
!
Canarias
!
Guadeloupe Martinique
!
!
Réunion
!
Year: 2012, except; BE (2011), DE,
EL, NL, (2010), FR, UK, (2009) TR,
(2006), PT (2005).
Helsinki
!
Tallinn
Oslo
!
!
Stockholm
Guyane
!
!
Riga
!
Madeira
!
København
Vilnius
!
!
Minsk
!
Dublin
!
Acores
Warszawa
Berlin
Amsterdam
!
!
!
!
London
Kyiv
!
Region Level:
NUTS 0 - EE, CY, LV, LT, LU, MT,
IS, HR, TR
NUTS 1 - BE, EL, HU, PL, UK
NUTS 2 - BG, CZ, DK, DE, IE, ES,
FR, IT, NL, AT, PT, RO, SI, SK, NO,
CH
!
Bruxelles/Brussel
!
Praha
!
Luxembourg
!
Paris
!
Kishinev
Wien ! Bratislava
!
Budapest
!
!
Bern
!
Vaduz
!
Ljubljana Zagreb
!
Bucuresti
!
Beograd
Percentage population in households
with <60% of the national median
equivalised disposable income.
!
!
Sarajevo
!
Sofiya
!
Podgorica
!
Roma
Ankara
!
!
Tirana
!
Madrid
Skopje
!
!
2.7 - 9.9
Lisboa
!
10.0 - 14.9
Athina
!
Nicosia
!
El-Jazair
!
15.0 - 19.9
Tounis
!
!
Valletta
20.0 - 29.9
30.0 - 42.3
ARoP – At Risk of Poverty Rate (TiPSE - NUTS 3)
Reykjavik
!
Sources:
Canarias
!
Guadeloupe Martinique
!
!
Helsinki
!
Tallinn
Oslo
!
!
Stockholm
Guyane
!
!
Riga
!
Madeira
!
København
!
Minsk
!
!
Acores
Warszawa
Berlin
Amsterdam
DK, SE, FI, NO, IE, NL, FR, UK, HR National Statistical Institutes
LV, HU, RO, SI, SK - World Bank.
BG, CZ, EE, CY, LT, LU, MT, PL, LI Eurostat Regio Database (NUTS 0-2)
!
!
!
!
BE, DE, EL, ES, IT, AT, PT, TR, CH ESPON TiPSE project.
Vilnius
!
Dublin
London
Réunion
!
Kyiv
!
!
Bruxelles/Brussel
!
Praha
!
Luxembourg
!
Paris
!
Kishinev
Wien ! Bratislava
!
Budapest
!
Percentage population in households
with <60% of the national median
equivalised disposable income.
!
Bern
!
Vaduz
!
Ljubljana Zagreb
!
Bucuresti
!
Beograd
!
!
Sarajevo
!
Sofiya
!
Podgorica
!
Skopje
Roma
!
Tirana
!
Madrid
No Data
Ankara
!
!
< 9.9
!
Lisboa
!
10.0 - 14.9
Athina
!
Nicosia
!
El-Jazair
!
15.0 - 19.9
Tounis
!
!
Valletta
20.0 - 29.9
30.0 - 63.4
What can we learn from this map?
Income poverty rates are;
•higher in urban areas in 4 countries (the centre)
•higher in rural/intermediate areas in 11 countries (Med. and East)
•No clear U-R difference in 5 countries (mostly NW)
What socio-economic indicators seem to be associated with income poverty?
o
Regression across ESPON countries, EU15, EU12, Welfare regime groups,
P&SE clusters.
o
Some key results:
•
Unemployment (total, youth and long-term) is closely associated with
ARoP throughout Europe
•
Profiles of relationships in EU15 and EU12 are slightly different:
• EU15 – labour market characteristics, elementary occupations.
• EU12 – accessibility, primary sector, education & skills, productivity.
•
GDP per capita
• strong negative correlation in EU12 (i.e regional performance drives
poverty rates), BUT
• weak relationship in EU15 (intra-regional distributional effects more
important than inter-regional variation in overall performance)
ARoP – Can we improve on it?
1. 28 different poverty lines…?
2. EU-SILC – sample size issue:
• regression modelling is a problematic short term fix
• longer term solution is to develop EU-wide register data.
Eurostat “Register Hub”?
3. Income is only one aspect of poverty, living costs also very substantially
Housing cost adjustment is not enough (urban bias)
National
Medians
Reykjavik
Reykjavik
!
!
Canarias
!
Guadeloupe Martinique
!
!
National
Quintiles
!
Canarias
!
Guadeloupe Martinique
Helsinki
Réunion
Guadeloupe Martinique
!
!
!
!
Tallinn
Oslo
Guyane
!
!
!
Guyane
!
!
Tallinn
Oslo
!
Stockholm
!
!
Stockholm
Guyane
!
!
Riga
!
Riga
Madeira
Riga
!
!
!
Madeira
Madeira
!
København
Vilnius
!
!
København
Minsk
!
København
Vilnius
!
!
!
Vilnius
!
Minsk
!
Minsk
!
Dublin
!
Dublin
!
!
Helsinki
!
Tallinn
Stockholm
!
Helsinki
!
!
Canarias
!
Réunion
!
!
Oslo
Median
= 100
Reykjavik
Dublin
!
!
Acores
Amsterdam
Acores
!
!
!
!
London
Warszawa
Berlin
Kyiv
!
!
!
London
!
Warszawa
Berlin
Amsterdam
!
Acores
!
!
Bruxelles/Brussel
!
!
Praha
!
Praha
!
Luxembourg
!
Paris
!
Paris
!
!
!
Kishinev
Wien ! Bratislava
!
Budapest
!
!
!
Luxembourg
!
Paris
Kishinev
Wien ! Bratislava
!
Budapest
Bern
!
Bruxelles/Brussel
!
Praha
Luxembourg
!
Bern
!
Bern
!
!
Ljubljana Zagreb
!
Beograd
!
Vaduz
!
Ljubljana Zagreb
!
Bucuresti
!
Beograd
!
Sofiya
Sarajevo
!
Roma
Ankara
Podgorica
!
!
Tirana
!
Sofiya
!
Skopje
!
Madrid
!
Sofiya
!
!
Roma
!
!
Skopje
Podgorica
Ankara
!
!
!
!
!
Athina
Athina
!
Athina
!
Nicosia
!
Nicosia
!
Nicosia
!
Tounis
El-Jazair
!
!
Valletta
!
!
!
Lisboa
!
!
Ankara
Tirana
!
Madrid
!
Lisboa
!
Skopje
!
Roma
Tirana
!
Madrid
!
Lisboa
El-Jazair
!
!
Sarajevo
!
Podgorica
Bucuresti
!
!
Beograd
!
Sarajevo
!
!
Vaduz
Bucuresti
!
Kishinev
Wien ! Bratislava
!
Budapest
!
!
Vaduz
Ljubljana Zagreb
!
Kyiv
!
Bruxelles/Brussel
!
!
!
!
!
London
!
Warszawa
Berlin
Amsterdam
Kyiv
!
!
El-Jazair
Tounis
!
!
!
Valletta
Tounis
!
!
Valletta
Réunion
!
From Poverty to Social Exclusion…?
Social Exclusion is relational, multi-dimensional and dynamic.
In TiPSE we identified four broad domains of social exclusion:
•
•
•
•
Earning a living (income; employment)
Access to Services (health; education; housing; transport ; communications)
Social Environment (Age; Ethnic Composition; Migrants; Crime and Safety)
Political participation (Citizenship; Voice)
Such a complex multi-faceted concept cannot be mapped by a single
indicator. Moreover social exclusion is a set of processes rather than
static characteristics. We identified proxy indicators which reflect the
risk of different kinds of exclusion.
Mapping SE Domains…
Mapping SE Domains…
Policy Analysis - EU
EU Social Policy and the ‘Domains’ of SE
• EU Social Policy emphasises the ‘earning a living’ domain with less
attention given to the other domains of ‘access to basic services’,
‘social environment’ and ‘political participation’.
• All EU policies should contribute to economic, social and territorial
cohesion but it is not clear how social strategies seek to address
territorial cohesion below the MS level. Social groups at risk of P&SE
are targeted but not regions at risk. As a result, many EU policies are
‘territorially blind’.
• The targeting of regions for allocation of ESIF is according to
GDP/head but this is poorly correlated with measures of P&SE
Policy Analysis – Member States
Welfare Regimes:
• Society-based
• Individual
• State-based
• Familial
• Transitional
Governance Issues:
• Lack of data
• Subsidiarity
• Coordination
• Empowerment
Policy Implications
Conceptualising P&SE
Recommendation:
• P&SE are closely related
but nevertheless distinct.
• The 5 aspects of P&SE are
differentially addressed in
both data assembled and
in policy development:
• Use broad conceptualisation
of P&SE in all policy arenas and
provide appropriate data to
support this…
• We would commend the
’domains’ of P&SE developed in
TIPSE as a useful tool in this
respect.
•
•
•
•
•
At risk of poverty rate
Earning a living
Access to services
Social environment
Political participation
Policy Implications
Issues of Scale:
• The territorial risk of P&SE
varies according to scale.
• Different clusters are
associated with high incidence
of P&SE from different TIPSE
domains - as also are different
scales.
• Identification of P&SE in
social and political domains,
and at sub-national level, is
underplayed by social OMC.
Recommendations:
• P&SE data should be collected
and analysed regularly at least
at NUTS2 (ideally NUTS3)
• Elaborate geography of P&SE.
• Assess value of other
methods
• Consider producing a report
on territorial aspects of P&SE
Policy Implications
Geographies of P&SE:
Why do these territorial
differences emerge?
•Rural – urban differences:
• Poverty risk worse in rural areas
in poorer countries, but not rich.
• SE domains differ by R-U.
•Regions of immigration.
• Border regions
• Parts of cities
•Macro-level effects
• N-S and E-W splits in EU
• Regional splits within countries
Recommendations:
• Use TIPSE insights to build
territorial cohesion dimension
into all policies, eg CAP,
URBACT.
• Recognise that geographies of
P&SE reflect processes at
several scales; and these call
for policy at multiple levels
Policy Implications
Effects of the Crisis:
Not a major focus of TIPSE, but…
• Most Mediterranean states, Baltic states, Ireland and some
East Central European countries were worst hit.
• Widespread austerity measures of MSs led to cuts in state
services which reinforced P&SE, especially in remoter areas.
• Intra-Europe migration patterns changed because of the Crisis
• In some regions the outmigration of young people has made
the challenge of an ageing population more acute.
• The capacity of family-based support structures in southern
Europe appears further stretched by the Crisis
Policy Implications
Recommendations:
• National Reform Programmes should
include Regional Chapters
• CLLD/LEADER type approach should
be encouraged as a means of
alleviating P&SE at local level.
• In resource & policy targeting, high
risk of P&SE should be considered as
well as GDP/head.
• Differential living costs should also be
reflected in indicators.
• Common issues could usefully be
discussed among similar regions
• Consider how best to collate and
disseminate case studies
Monitoring recommendations:
• Broaden set of data expressing
multidimensionality of SE
• Data accessible at as low NUTS level
as possible
• Encourage mutual learning among
monitoring bodies, through
secondments and exchange of
experiences
• Making existing data (indicators)
accessible in a more user-friendly
manner.