Welfare State Reform in Southern Europe

Welfare State Reform in Southern Europe analyses policies against poverty and social exclusion in Italy,Spain, Portugal and Greece in the wider framework of welfare state modernization in these countries,focusing especially on developments since the 1990s.

The specific focus is on innovations, experiments and debates relating to the guarantee of a minimumincome to individuals and households lacking sufficient resources. The chapters on Italy, Spain, Portugaland Greece are organized around a common structure and include discussion of:

• Welfare state development in a historical perspective.• Poverty trends and anti-poverty policies.• Debates and initiatives in the field of minimum income.

The book includes a comparative introduction, which highlights the features of Southern European welfarestates and an additional chapter illustrates developments occurring in the countries of Eastern and SouthEastern Europe, discussing to what extent the experience of Southern Europe can be a useful point ofreference for these countries.

This book will interest students and researchers in the field of welfare states and public policy.Maurizio Ferrera is Professor of Social Policy at the University of Milan. He is also Deputy Director of

the Centre for Comparative Political Research at Bocconi University and Director of the Research Unit onEuropean Governance (URGE) at the Collegio Carlo Alberto of Moncalieri, near Turin.

Routledge/EUI studies in the political economy of welfareEdited by Martin Rhodes and Maurizio Ferrera

The European University Institute, Florence, ItalyThis series presents leading edge research on the recasting of European welfare states. The series isinterdisciplinary, featuring contributions from experts in economics, political science and social policy. Thebooks provide a comparative analysis of topical issues, including:

• reforms of the major social programmes—pensions, health, social security• the changing political cleavages in welfare politics• policy convergence and social policy innovation• the impact of globalization.

1 Immigration and WelfareChallenging the borders of the welfare stateEdited by Michael Bommes and Andrew Geddes

2 Renegotiating the Welfare StateFlexible adjustment through corporatist concertationEdited by Frans van Waarden and Gerhard Lehmbruch

3 Comparing Welfare CapitalismSocial policy and political economy in Europe, Japan and the USAEdited by Bernhard Ebbinghaus and Philip Manow

4 Controlling a New Migration WorldEdited by Virginie Giraudon and Christian Joppke

5 The Moral Economy of Welfare StatesBritain and Germany comparedSteffen Mau

6 Welfare State Reform in Southern EuropeFighting poverty and social exclusion in Italy, Spain, Portugal and GreeceEdited by Maurizio Ferrera

Welfare State Reform in SouthernEurope

Fighting poverty and social exclusion in Italy, Spain,Portugal and Greece

Edited by Maurizio Ferrera

LONDON AND NEW YORK

First published 2005by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

Simultaneously published in the USA and Canadaby Routledge

270 Madison Avenue, New York, NY 10016

Routledge is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2005.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go towww.eBookstore.tandf.co.uk.”

© 2005 Maurizio Ferrera for selection and editorial matter;individual contributors their chapters

All rights reserved. No part of this book may be reprinted orreproduced or utilized in any form or by any electronic, mechanical,

or other means, now known or hereafter invented, includingphotocopying and recording, or in any information storage or

retrieval system, without permission in writing from the publishers.

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication DataWelfare state reform in Southern Europe: fighting poverty andsocial exclusion in Italy, Spain, Portugal, and Greece/edited by

Maurizio Ferrera.p. cm.

Includes bibliographical references and index.1. Europe, Southern—Social policy. 2. Welfare state—Europe,

Southern. 3. Human services—Europe, Southern. 4. Marginality,Social—Europe, Southern. I. Ferrera, Maurizio.

HN650.7.A8W45 2005361.6�1�094–dc22

2004017566

ISBN 0-203-35690-X Master e-book ISBN

ISBN 0-203-66932-0 (Adobe eReader Format)ISBN 0-415-32409-2 (Print Edition)

Contents

List of illustrations vi

List of contributors ix

Acknowledgements x

List of abbreviations xi

1 Welfare states and social safety nets in Southern Europe: an introductionMAURIZIO FERRERA

1

2 Greece—fighting with hands tied behind the back: anti-poverty policy without aminimum incomeMANOS MATSAGANIS

24

3 Italy—striving uphill but stopping halfway: the troubled journey of the experimentalminimum insertion incomeSTEFANO SACCHI AND FRANCESCA BASTAGLI

65

4 Spain—poverty, social exclusion and “safety nets”ANA ARRIBA AND LUIS MORENO

110

5 Portugal—a virtuous path towards minimum income?LUÍS CAPUCHA, TERESA BOMBA, RITA FERNANDES AND GISELA MATOS

163

6 Poverty and the safety net in Eastern and South-Eastern Europe in the post-communisteraDIMITRI A.SOTIROPOULOS

210

Index 233

Illustrations

Figures

1.1 The institutional location of RMI-type schemes in under-developed areas 183.1 Social benefits by group of functions 2001 (% of total social benefits) 713.2 “Other” social benefits 2001 (% of total social benefits) 713.3 Italy’s public expenditure on social protection benefits by category 2002 (%) 723.4 At-risk-of-poverty rate after social transfers, EU-15 2000 (ECHP 2001) 773.5 At-persistent-risk-of-poverty rate after social transfers, EU-15 2000 (ECHP 2001) 783.6 Municipalities involved in the first and second RMI experimentation phase 884.1 Poverty rates in Spain (ECHP 1994–97) 1204.2 The Spanish public “safety net”: levels and means of protection 1365.1 Evolution of social facilities by target 1805.2 Annual evolution of GMI beneficiaries over the total resident population (%) 1925.3 GMI annual beneficiaries and household representatives 1925.4 Reasons for withdrawal from GMI 1915.5 Insertion areas (%) 1935.6 Beneficiaries by age group (%) 1805.7 Situation of the beneficiaries of GMI in the labour market 1945.8 GMI types of households (%) 194

Boxes

2.1 Retirement benefits received by low-income groups 382.2 Family benefits received by low-income groups 402.3 Housing benefits received by low-income groups 412.4 Anti-poverty measures introduced in 2000–02 43

Tables

1.1 Selected social indicators for Southern Europe 51.2 Minimum income programmes (2000) 132.1 Total expenditure on social protection (1990–99) 302.2 Distributional impact of social transfers (1999) 31

2.3 Expenditure on social security benefits (2001) 322.4 Relative poverty rate (1981–99) 332.5 Relative poverty gap (1981–99) 332.6 Relative poverty profiles (1999) 342.7 Absolute poverty (1981–99) 352.8 Extreme poverty profiles (2000) 362.9 Annual benefit rates (2002) 2.10 Selected anti-poverty benefits in Greece (2002) 452.11 Design features of selected benefits 492.12 Inconsistencies of income tests: EKA� 502.13 Indexation and revaluation of selected benefits (1996–2001) 502.14 Income thresholds as percentage of the poverty line (2002) 512.15 Effects of minimum income under alternative scenarios (2000) 553.1 Headcount ratios (households) for relative and absolute poverty (1997–2002) 733.2 The Carbonaro Equivalence Scale 733.3 Headcount ratio and average poverty gap ratio (relative poverty) (1997–2002) 743.4 Relative poverty by geographic area (2002) 743.5 Distribution of poor as opposed to residents by geographic area in 2002 (%) 743.6 Relative poverty rates (%) by geographic area and family characteristics, 2002 753.7 Relative poverty rates (%) by geographic area and professional situation of the person of

reference (2002) 75

3.8 Absolute poverty (headcount ratio and average poverty gap ratio) (1997–2002) (%) 763.9 Incidence and distribution of absolute poverty by geographical area (2002) (%) 773.10 Absolute poverty incidence by family size and type (2002) (%) 783.11 Risk of poverty for households with children relative to that in other selected family types

(ratios), 1995 and 1993–95 for data on persistence 78

3.12 Percentage of expenditure on selected poverty alleviation measures going to various richestpopulations

83

3.13 Targeting indicators of anti-poverty measures in Italy 833.14 RMI threshold in 2000 according to household size (in euros) 893.15 RMI claimant and beneficiary households (1998–2000) 913.16 RMI beneficiary households and individuals on 31 December 2000 913.17 Composition of beneficiary households on 31 December 2000 923.18 RMI beneficiaries: status in relation to labour market on 31 December 2000 933.19 Beneficiaries involved in insertion programmes on 31 December 2000 933.20 Distribution of activated beneficiaries among insertion programme types 943.21 RMI households and the reported exit from the RMI, reasons for ending benefit 973.22 Chronology of legislative provisions in the field of poverty alleviation 1004.1 Poverty rates by individual and household characteristics (ECHP 1996) 1214.2 Territorial distribution of poverty in Spain (ECPF 1998) 1234.3 Safety net benefits and minimum schemes: amounts and beneficiaries (2000) 1254.4 Main features of safety net benefits (2000) 1284.5 National Action Plans’ aggregate budgetary efforts (2001–04) (� million) 1384.6 Typology of the regional programmes of minimum income of insertion 1424.7 Regional expenditure and coverage of the programmes of minimum income schemes (2000) 144

vii

41

4.8 Minimum income schemes: main legislative dispositions 1494.9 Evolution of the regional expenditure on minimum income schemes 1514.10 Beneficiaries of regional minimum income schemes (households) 1515.1 Social protection in Portugal—historical overview (before 25 April 1974) 1665.2 Lei de Bases da Segurança Social (1984) 1715.3 Poverty rates in Portugal by relevant features, 1980/81 and 1989/90 1725.4 Poverty and income distribution in Portugal (%) 1755.5 At-risk-of-poverty rate by household type in Portugal (%) 1755.6 At-risk-of-poverty rate by activity status and income source in Portugal (%) 1765.7 Main employment indicators 1785.8 Social protection expenditure (% of GDP) 1815.9 Main aspects of GMFs original design 1875.10 GMI expenditure 1895.11 Chronology of legislation on poverty alleviation in Portugal 1785.12 Main benefits in the social protection system in Portugal (2000) 2025.13 Main schemes in the social protection system in Portugal (2000) 2046.1 Income inequality in Eastern and South-Eastern Europe, before and after the transition 2136.2 Absolute poverty rates and GNP in dollars per capita (1998) 2156.3 Sub-regional trends in social welfare (% change 1989–96) 2176.4 Earning inequality (GINI coefficient) and extent of poverty affecting total population in Bulgaria

and Romania (%) 219

viii

Contributors

Ana Arriba is a postdoctoral researcher at the Spanish National Research Council (UPC-CSIC) inMadrid.

Francesca Bastagli is an economist and has worked for the Poverty Reduction Group at the World Bank.She is currently at the London School of Economics.

Teresa Bomba is a Research Fellow in the Department of Studies, Statistics and Planning of the Ministryof Social Security and Labour, Lisbon.

Luís Capucha is Assistant Professor in the Department of Sociology at the Institute of Social Sciencesand Business Studies (ISCTE) and Senior Researcher at the Centre for Research and Studies in Sociology(CIES) in Lisbon.

Rita Fernandes is an economist in the Department of Studies, Statistics and Planning of the Ministry ofSocial Security and Labour, Lisbon.

Maurizio Ferrera is Professor of Social Policy at the University of Milan. He also directs the Centre forComparative Political Research “Poleis” at Bocconi University, Milan, and the Research Unit onEuropean Governance (URGE) at the Collegio Carlo Alberto of Moncalieri, Turin.

Gisela Matos is a sociologist at the Institute of Solidarity and Social Security of the Ministry of SocialSecurity and Labour, Lisbon.

Manos Matsaganis is Assistant Professor in the Department of International and European Economics atthe Athens University of Economics and Business.

Luis Moreno is Senior Research Fellow at the Spanish National Research Council (CSIC) in Madrid.

Stefano Sacchi is Research Fellow at the Centre for Comparative Political Research “Poleis” of BocconiUniversity, Milan, and Scientific Coordinator of the Research Unit on European Governance (URGE) atthe Collegio Carlo Alberto of Moncalieri, Turin.

Dimitri A.Sotiropoulos is Assistant Professor in the Department of Political Science and PublicAdministration of the University of Athens.

Acknowledgements

This volume is the final product of a research project on “Fighting poverty and social exclusion in SouthernEurope”, originally funded by the European Commission (FIPOSC, contract HPSE-CT-2001–60020). Theproject began in 2002 and was co-ordinated by the Centre for Comparative Political Research “POLEIS” atBocconi University, Milan, with the participation of the University of Crete, the Spanish National ResearchCouncil (IESA-CSIC) and the Department of Studies, Prospective and Planning, Ministry of Labour andSolidarity, Portugal. A special thanks to Francesca Bastagli, who contributed not only to the research onItaly, but also to the overall project organization. The first report of the project was presented and discussedat an international seminar held at ISPI (Istituto per gli Studi di Politica Internazionale), Milan, in May2002. We thank all the participants to that seminar for the useful comments received. We also warmly thankthe Compagnia di San Paolo for the valuable financial support offered to complete this volume.

Abbreviations

CCOO Comisiones Obreras (Spanish Trade Union)

CERMI Comité Español de Representantes de Minusválidos (Spanish Committee ofRepresentatives of Handicapped People)

CLAS Comissões Locais de Acção Social (Local Commissions for Social Action)

CLES Centro di Ricerche e Studi sui Problemi del Lavoro, dell’Economia e dello Sviluppo(Italian Research Centre on Labour, Economy and Development)

CNRM Comissão Nacional do Rendimento Mínimo (National Commission for MinimumIncome)

CSF Comissões Sociais de Freguesia (Parish Social Committee)

DGV Directorate General for Employment and Social Affairs

EC European Community

ECHP European Community Household Panel

ECPF Encuesta Continua de Presupuestos Familiares (Spanish Family Accounts ContinuousSurvey)

EDIS Equipo de Investigación Sociológica (Consulters on Sociological Research)

EEC European Economic Community

EKAS Pension Social Solidarity Supplement

EMU European Monetary Union

EPF Encuesta de Presupuestos Familiares (Spanish Family Accounts Survey)

ESF European Social Fund

EU European Union

FAINA Fundo de Apoio a Inserção em Novas Actividades (Support Fund for Insertion in NewActivities)

FIPOSC Fighting Poverty and Social Exclusion in Southern Europe

FONAS Fondo Nacional de Asistencia Social (Spanish National Fund on Social Assistance)

FSU Former Soviet Union

GDP Gross Domestic Product

GMI Guaranteed Minimum Income

HBS Household Budget Survey

IDS Instituto para o Desenvolvimento Social (Institute for Social Development)

IKA Social Insurance Foundation

ILO International Labour Office

INE Instituto Nacional de Estadística (Spanish National Institute of Statistics)

IMSERSO Instituto de Migraciones y Servicios Sociales(Spanish Public Institute on Migration andSocial Services)

INPS Istituto Nazionale della Previdenza Sociale (Italian National Institute for SocialInsurance)

INSERSO Instituto de Servicios Sociales (Spanish Public Institute on Social Services)

IRS Istituto per la Ricerca Sociale (Institute for Social Research)

ISE Indicatore della Situazione Economica (Indicator of Socio-Economic Situation)

ISEE Indicatore della Situazione Economica Equivalente (Indicator of Equivalent Socio-Economic Situation)

ISPL International Standard of Poverty Line

ISTAT Italian National Statistical Institute

IU Izquierda Unida (Spanish Left-wing Political Coalition)

LISMI Ley de Integración Social de los Minusválidos (Spanish Handicapped Social IntegrationLaw)

MIG Minimum Income Guarantee

MSE Mercado Social de Emprego (Social Employment Market)

MTAS Ministerio de Trabajo y Asuntos Sociales (Spanish Ministry of Labour and Social Affairs)

NAP National Action Plan

NAPSI National Action Plan for Social Inclusion

NGO Non-Governmental Organization

OAEE Insurance Organization for the Self Employed

OAE Manpower Employment Organization

OECD Organization for Economic Cooperation and Development

OEK Workers Housing Organization

OGA Agricultural Insurance Organization

OMC Open Method of Coordination

PACI Patto di Aiuto Concordato e Individualizzato (Pact of Negotiated and IndividualizedHelp)

NAPincl Plano Nacional de Acção para a Inclusão (National Action Plan for Inclusion)

PNC Pensiones no Contributivas de la Seguridad Social (Spanish Social Security Non-contributory Pensions)

PNLCP Programa Nacional de Luta contra a Pobreza (National Programme to Fight Poverty)

PP Partido Popular (Spanish Popular Party)

PPS Purchasing Power Standard

PSD=PPD/PSD Partido Social Democrata (Portuguese Socialdemocratic Party)

xii

PSOE Partido Socialista Obrero Español (Spanish Socialist Party)

RESSAA Regime Especial de Segurança Social das Actividades Agrícolas (Special Social SecurityScheme for Agriculture)

RMG Rendimento Mínimo Garantido (Minimum guaranteed income)

RMI Reddito Minimo di Inserimento (Minimum insertion income)

RMI Revenu Minimum d’Insertion (Minimum insertion income)

RSI Rendimento Social de Inserção (Social Insertion Income)

RSSV Voluntary Social Security Scheme

RUI Reddito di ultima istanza (Last Resort Income)

SOE Seguro Obligatorio de Enfermedad (Spanish Sickness Compulsory Insurance)

SOVI Seguro Obligatorio de Vejez e Enfermedad (Spanish Old age and Sickness CompulsoryInsurance)

TAPSI Planes de Acción Territoriales para la Inclusion Social (Territorial Action Plans for SocialInclusion)

TFR Trattamento di fine rapporto (End of Contract Payment)

UGT Union General de Trabajadores (Spanish Trade Union)

xiii

1Welfare states and social safety nets in Southern Europe

An introduction

Maurizio Ferrera

Welfare states and safety nets

The metaphor of the “safety net” was first coined by acrobats to designate a large, strong and flexible pieceof fabric ready to catch them if they fell or jumped from their trapeze. Later it started to be used in thesphere of financial transactions and insurance markets in order to indicate clauses of basic guarantee forinvestors, often covered by legal contracts. In both cases, the metaphor suggests a sense of protection: frominvoluntary accidents, in the first case; from the unforeseeable hazards of markets, in the second. It alsoevokes the idea of a commitment: the pledge (a legal one, in the case of financial contracts) of a minimumsecurity against potential failures.

In the past few decades, this evocative metaphor has been adopted by international social policy debates.The establishment of “social safety nets” has started to be advocated as a priority goal for many developingcountries, in the framework of aid programmes sponsored or funded by international organizations such asthe World Bank (Deacon 2000; Vivian 1995; World Bank 2000). In this context, the metaphor has beenused to designate a set of compensatory measures meant to mitigate the shortterm negative effects of structuraladjustment policies or severe conjunctural crises. But the term “social safety net” has also been increasinglyused with reference to the experience of Organization for Economic Cooperation and Development (OECD)countries, as a possible novel brand of traditional social assistance policies in general and for means-tested,“last resort” schemes in particular, such as minimum income schemes.1 These schemes do in fact respond tosituations of “accidental fall”, often linked to market hazards, by offering basic protection to individuals andhouseholds as a matter of right (the “guarantee” aspect).

The link between the safety net notion and guaranteed minimum income schemes has grown particularlystrong in Europe, in the wake of two distinct developments: (1) the emergence of a new generation of socialassistance schemes at the national level, on the one hand; and (2) the increasing role and activism of theEuropean Union in the social policy realm, on the other.2

As is well known, the modern welfare state emerged in Europe out of a long-standing tradition oflocalized and discretionary poor relief, codified in Northern Europe under specific “poor laws” (Alber1982; De Swaan 1987). With the birth, consolidation and expansion of social insurance (whether of theuniversalistic, “Beveridgean” sort, or the occupational, “Bismarckian” sort), traditional ad hoc interventionsof public relief gradually lost functional and financial salience in most countries, especially during the so-called Trente Glorieuses. Between the 1960s and 1970s, however, a new generation of social assistanceschemes and benefits made their appearance, with two main objectives: (1) to fill the coverage gaps at themargins of the extant social insurance programs, in particular for some categories of people and for some

new types of need; and (2) to establish a minimum guarantee—a safety net, precisely—for the wholecitizenry, below which nobody would be allowed to fall.

The novelty of these new schemes and benefits in respect of the discretional social assistance of the pastlies in their right-based nature. Protection is offered to individuals, conditional upon the assessment of amanifest need (selectivity) and on the impossibility of these individuals to cope with such need with theirown resources (residuality). Eligibility is not linked to prior contributions, and funding typically comes fromgeneral revenues. Despite selectivity and residuality, the protection offered rests on subjective andjudiciable rights and on codified administrative procedures, which orient and constrain public decisions onindividual cases.

The poor elderly were among the first category to be targeted by this new generation of policies, throughthe establishment of “social pensions” open to all the aged who could not access ordinary, insurance-basedbenefits. The long-term unemployed, the disabled with special needs, single mothers, needy students havebeen other prominent social groups for which categorical schemes of means-tested social assistance havebeen set up in various countries. As regards the general “safety net”, the UK pioneered developments onthis front by establishing a National Assistance scheme as early as in 1948 (subsequently re-labelled IncomeSupport). But most countries have followed suit, introducing some sort of generalized minimum incomeguarantee between the 1960s and the 1990s, as a means of filling the gaps left by existing means-testedcategorical schemes and thus contrasting situations of extreme indigence (Eardley et al. 1996; Guibentif andBouget 1997).

The second development has been the increasing prominence of the issues of poverty and socialexclusion on the agenda of the European Union and the Commission in particular. This development startedduring the 1980s, with the launch of three subsequent Poverty Programmes and the establishment, in 1990,of an Observatory on National Policies to Combat Social Exclusion. The turning point was, however, 1992:in that year, Recommendation 92/441 officially recognized that “social exclusion and the risk of povertyhave become more prevalent and more diversified over the last 10 years” and, more importantly, invited themember states to institutionalize “the basic right of a person to sufficient resources and social assistance”, inorder to be able to live in a manner “compatible with human dignity”. The presence of a generalized safetynet was, in other words, indicated to be an essential ingredient of the European social model—in all itsnational institutional variants. In 1997, with the Treaty of Amsterdam, the fight against exclusion became anofficial objective of the EU and “the integration of persons excluded from the labour market” became anarea in which the Community could support and complement the activities of the member states (Art. 137).In the context of the Lisbon strategy, in December 2000 a “social inclusion process” was eventuallylaunched, based on the open method of coordination, aimed at promoting four ambitious objectives (knownas the Nice objectives): (1) to facilitate participation in employment and access by all to the resources,rights, goods and services; (2) to prevent the risk of exclusion; (3) to help the most vulnerable; and (4) tomobilize all the relevant bodies. The launch of the social inclusion process has enhanced the symbolicvisibility and the institutional salience of anti-poverty policies— and of minimum income schemes inparticular—at both the supranational and national levels (Ferrera et al. 2002), underlining the importancenot only of cash transfers, but also of accompanying measures favouring social and labour marketintegration.

The Southern European countries have played an active part in promoting the EU agenda on poverty andsocial exclusion. This is especially true of Portugal, which has played a pivotal role in the launch of theinclusion process during its Presidency in 2000. At the domestic level, however, policy innovation has beenrather difficult in this area. For Italy, Greece, Portugal and Spain, the building of a safety net at the bottomof their welfare edifices has been (and still partly is) a very demanding challenge, for reasons which have to

2 MAURIZIO FERRERA

do with their socio-economic contexts, social and political cultures and the institutional configuration oftheir social policy regimes. Yet important reforms or at least experiments have been carried out in the pastdecade, which have created a more fertile ground for the consolidation of minimum income schemes andthus a more effective fight against poverty and exclusion (Moreno 2003).

The aim of this volume is to illustrate and discuss the Southern European path towards the establishmentof a generalized social safety net within the welfare states of Greece, Italy, Spain and Portugal in the widercontext of the traditional policies of social assistance and targeted benefits. The specific focus will be onreforms, experiments and debates with regard to the guarantee of a minimum income to individuals andhouseholds lacking sufficient resources. The experience of each country will be covered by a distinctchapter, organized around a common grid: welfare state development from a historical perspective; povertytrends and anti-poverty policies; and debates and initiatives in the field of minimum income. An additionalchapter will illustrate developments occurring in the countries of Eastern and South Eastern Europe. Thischapter will also discuss to what extent the experience of Southern Europe can provide useful “lessons” forthese countries. The book is the final product of a research project on “Fighting Poverty and SocialExclusion in Southern Europe”, funded by the European Commission under the Fifth FrameworkProgramme.3 In the remainder of this Introduction, I will outline the overall context of our project, startingwith a brief presentation of the features and problems of Southern European welfare, with specific referenceto social assistance, then summarizing developments in the four national cases and finally drawing somecomparative conclusions.

Southern European welfare and the delayed development of social assistance

The distinctiveness of Southern Europe with respect to other macro-areas of the continent has beenhighlighted by a rich historical and social science literature.4 The nations of Southern Europe have followeda specific path to modernization (in the broad sense of the concept) and still share a number of commontraits in their cultural backgrounds and political economies. There are, of course, significant differencesbetween the four countries of the region: the intra-area variation is certainly greater than in the Nordiccontext, though probably lower than in Central Continental Europe. It would be difficult to deny that thenotion of “Southern Europe” has not only a geographical, but also a substantive, cultural and politico-economic connotation.

The idea that this area is distinct also as regards social policy and the labour market has come to bewidely shared in recent debates.5 We can try to capture this distinctiveness by referring to the three grandachievements accomplished (or at least pursued) by the various OECD countries during the three decadesafter the Second World War (Scharpf and Schmidt 2000), i.e.:

1 full employment with “good jobs” for all (men) who were expected to work for a living;2 social insurance of workers against the risk of sickness, invalidity, unemployment and old age, coupled

with generous family benefits;6

3 social assistance to prevent the poverty of those without other sources of support.

On each of these three crucial fronts, Southern Europe has indeed attempted to move in the directionfollowed by the other, more advanced countries. But this effort has encountered great obstacles and thepolicy solutions that were put in place in the 1960s and 1970s have occasionally worked to exacerbate,rather than overcome, these very obstacles.

INTRODUCTION 3

As far as the first objective is concerned, i.e. the promotion of an inclusive “Fordist” labour market, thefour Southern European countries have lagged chronically behind, both in terms of employment levels andin terms of an adequate supply of good jobs for those in employment.7This is partly due to a difference instarting conditions: in the 1940s and 1950s these countries were still predominantly based on agricultureand self-employment, with a very high incidence of the informal economy. The transition to industrialFordism was much more complex than elsewhere and was still under way when the oil shocks hit the Westerneconomies in the 1970s, increasing their vulnerability and structurally undermining the viability of Fordistarrangements as such. A number of institutional choices made during the Southern European transition toFordism, however, set these countries on a somewhat distinctive labour market route, characterized by apronounced insider-outsider cleavage. The highly protective employment regimes put in place for thoseworking in the core sectors of the economy (e.g. the public sector and large industrial enterprises) operatedas differentiating devices, gradually segmenting the Southern European labour markets into threejuxtaposed sectors, hugely different in terms of working conditions and job security: (1) the “regular”; (2)the “irregular” or “peripheral”; and (3) the “underground” sectors (Ferrera 1996; Moreno 2000; Peréz-Diazand Rodriguez 1994). The size and distinctiveness of the latter two sectors and the ensuing degree ofpolarization between guaranteed and non-guaranteed workers have seen no parallel elsewhere in Europe.

As far as social insurance is concerned, the four Southern European countries did proceed with theestablishment and consolidation of the standard social insurance programmes but, again, following a ratherdistinct path (Ferrera 1996; Rhodes 1997). To begin with, social insurance co-evolved with a segmentedlabour market, and thus acquired its own degree of internal polarization: generous entitlements for theregulars, modest benefits for the irregulars, and only meagre crumbs (if anything at all) for those workerswho were unable to establish any formal contact with the regular labour market. In addition, SouthernEuropean social insurance over-privileged from the very beginning the risk of old age (at least in terms oflegal formulas) and reserved only a marginal role for family benefits. In line with the conservative-corporatist tradition, it also privileged transfers over services and promoted the proliferation ofdifferentiated schemes, tied to occupational status. It has to be noted, however, that the link with theBismarckian tradition was broken in the field of health care (Guillén 2002). All four countries establishednational health services during the 1970s and 1980s: a rather rare but very significant instance of “pathshift”, which has worked to smoothen the profile of the welfare state edifice and has contributed toupgrading the allocative and distributive efficiency of social expenditure in these countries.

Social assistance and the fight against poverty have been, finally, the weakest front of achievement for ourfour countries—at least up to the late 1980s. The Southern European safety net evolved slowly, through asequence of fragmented and mainly categorical additions (orphans, widows, disabled, poor elderly, etc.)with disparate rules and differentiated benefit amounts, with a low (if not altogether lacking) degree ofintegration between cash benefits and services and with wide holes in the overall fabric of public provision.It is therefore not surprising that poverty levels after transfers remain very high in Southern Europe: 22 percent in Greece, 20 per cent in Italy and Portugal, 19 per cent in Spain, as against an EU average of 16.8 percent (excluding Southern Europe) (see Table 1.1). As we shall explain below, the worst implications of suchhigh poverty levels have been traditionally alleviated by strong ties of microsolidarity within the extendedfamily—a defining trait of Southern European social life.

Poor anti-poverty performance is partly linked to the limited reach of those in a situation of economicneed: in Greece and Italy, where the problem is the most serious, only about one-third of people in thelowest income quintile typically receive “social benefits other than pensions” (Marlier and Cohen-Solal2000). Many poor households are ineligible for social assistance because they fail to fulfil the narrowconditions set out by the various categorical programmes. Needless to say, those affected by this syndrome

4 MAURIZIO FERRERA

include the largest groups of outsiders: the long-term unemployed, the new entrants into the labour markets,the irregulars and underground workers and—increasingly—immigrants.

Which factors explain the marginal role of social assistance within Southern European systems of socialprotection? Why have these countries paid little attention to the third grand objective of the (European)welfare state project? This is, of course, a central question: a detailed reconstruction of the dynamicsresponsible for this syndrome will be offered by each of the national chapters in this book. A summaryanswer can, however, be sketched at this stage.

The first factor that must be called into question is of a “developmental” nature. In all Bismarckiancountries social assistance has tended to remain the Cinderella of social policy: minimum income schemesonly saw the light of day in the 1970s and 1980s; Southern European welfare states are still in maturationand will eventually catch up—the argument goes—also in this field. Even though it contains a grain of truth,this developmental explanation is not fully adequate. The relative neglect of a comprehensive anti-povertydimension in Southern Europe has in fact coincided with a steady growth of total social spending: ifanything, as mentioned, certain social programmes (most notably pensions, especially in Italy and Greece)are over-developed, to the point of crowding out investments in other policy areas. Therefore, it iselsewhere that the reasons for the low profile of social assistance in Southern Europe must be

Table 1.1 Selected social indicators for Southern Europe

Year Greece Italy Portugal Spain EU-15 EU-11

GDP per capita (� thousand) 2001 12.0 21.0 11.9 16.2 23.2 26.8

Social expenditure (% GDP) 1999 25.5 25.3 22.9 20.0 27.6 27.0

Poverty before social transfersa 1998 23.0 23.0 27.0 25.0 26.4 27.4

Poverty after social transfers 1998 22.0 20.0 20.0 19.0 17.7 16.8

Distribution of income (S80/S20)b 1998 6.5 5.9 7.2 6.8 5.4 4.4

People in jobless households 2000 4.2 5.0 1.2 5.1 4.5 3.7

Unemployment rate 2000 11.1 10.5 4.1 14.1 8.2 5.8

Youth unemployment rate 2000 29.6 30.8 8.9 26.2 16.2 11.3

Female employment rate 2000 41.2 39.6 60.3 40.3 54.0 60.3

Source: CEC (2002).Notesa Other than pensions,b Ratio of income shares earned by the top and bottom quintiles.

sought: in the unique set of constraints that have inhibited its development. The three most relevant of theseare the role of the family, the incidence of the irregular and underground economy and low administrativecapacities, especially at the peripheral, street-level end of the state apparatus.

As has been highlighted by a vast literature, Southern European families historically functioned as aneffective (though informal) safety net: a social “shock absorber” and welfare broker for their members,active across a whole range of policy fields such as child care, unemployment assistance, care for theelderly and the disabled or housing (Moreno 2004; Naldini 2003). The extended household, comprisingthree or more generations (and/or lateral kin), has survived for a long time, virtually up to the present: evenif actual co-habitation has been sharply declining in recent decades, the intensity of contacts and of solidaristicties among family members here is still much stronger than in all other parts of Europe. The proportion ofpeople aged above 16 living with their parents is in turn very high and there is evidence that resource

INTRODUCTION 5

pooling has indeed intensified in the 1990s (Fernández Cordón 1997). Southern European familialism isknown to heavily rely on unpaid female work. The “familialization” of social assistance functions has givenrise to a distinct gender regime (with formal and informal rules), treating women principally on the basis oftheir family roles as regards their duties and sending them unprotected in the market (especially in theirregular and underground sectors) in case of economic need (Saraceno 1994; Trifiletti 1999). Very often,caring for children or older relatives is only possible at the expense of erratic careers or full withdrawal fromthe labour market. The high incidence of strong and cohesive extended families and of a familistic culture(reflecting “axiologic-cultural” dynamics, to use Moreno’s words: see Chapter 4 in this volume) haveprobably played a blocking role, initially, in respect of public social assistance, keeping down the demandfor benefits and services in this field. Families would meet the care needs of all their members andguarantee basic economic security; the state could refrain from intervening on this front and concentrateinstead on other priorities (among which, typically, were pensions). Survey data for 2001 confirm that inSouthern Europe poor people have intense social contacts, especially with their relatives: more intense thanthe population as a whole in the same country and more intense than poor people in the other Europeancountries (Gallie and Paugam 2002). This syndrome has undoubtedly positive results in terms of inclusion:the poor remain more firmly integrated into the social fabric. But it has also generated some socio-economicpathologies: the low rates of female employment, especially in Spain, Italy and Greece, and partly also thedramatic decline of fertility among women— sandwiched between heavy home duties and unfriendly labourmarkets— clearly indicate the high social strains present in the Southern model of welfare (Moreno 2002).And if it is true that poor people in the South may suffer less from isolation and exclusion than in NorthernEurope, there are signs that family protection may also be perceived as a straitjacket (Gallie and Paugam2002:52).

The irregular and informal economy represents the second factor that —in combination with strongfamilialism—has worked to weaken both the functional need and the political demand for public anti-poverty interventions. The irregular economy has provided a substantial number of jobs for marginalworkers (e.g. seasonal workers in agriculture, the building sector or the retail trade), offering low wages—withlow contributions— but at least some anchor to the system of social protection (e.g. health care or minimumpensions). The informal economy has in its turn added an equally substantial range of earning opportunities,in the most disparate activities but with no formal link with the welfare state.8 The informal economy isestimated to produce between 15 and 30 per cent of total GDP in the countries of this area. To be sure, thisfact is by no means exclusive to Southern Europe: other European countries display sizeable informalsectors as well (Schneider and Enste 2000). However, at least two elements make the Southern Europeansituation distinct: (1) the sectoral and territorial concentration of informal activities (e.g. traditional sectors,including agriculture, small and very small businesses, home sub-contracted work, which in turn tend to bemore widespread in backward regions):9 and (2) the more pronounced separation between the formal andthe informal sector. While in countries such as Belgium, Sweden or Denmark (where the shadow economyhas been increasing since the 1980s) the informal sector primarily consists of (additional) undeclaredactivities on the side of people who have a formal job (“moonlighting”), in Southern Europe informalworkers tend to operate fully outside the formal circuit (“underground work”).10 The informal sectorconstitutes an important source of income for all the outsiders (especially women and young people).According to recent estimates, in Sicily, for example, in 2004 regular and irregular (but still “formal”)workers numbered ca. 1.4 million; underground workers numbered as many as 675,000.11 Poor families inSouthern Europe are known to pool all possible crumbs of income to make ends meet, for instance, thesocial pension of a grandfather, the “official” but modest seasonal earnings of the first spouse, the

6 MAURIZIO FERRERA

unemployment subsidy of the second spouse, topped by some undeclared income from the informal economy,occasionally brought home even by the so-called “economically active” children.12

While strong familialism and an extended irregular/underground economy have operated on the demandside of social protection, the third factor has operated on the supply side: low state capacities have restrainedinnovation and reform in the field of social assistance, for fear of activating or exacerbating particularisticbehaviours and syndromes of “group predation” (Arriba and Moreno, Chapter 4, this volume). As is wellknown, the delivery of means-tested benefits (such as, typically, a minimum income to poor claimants)requires administrative competencies that have been slow to develop in Southern Europe. Owing in part tothe legacy of authoritarianism, the administrative systems of this area have historically suffered from lowautonomy, low implementation effectiveness and weak (or in any case, belated) incentives to develop thoserelational and pragmatic skills which are necessary to manage individualized social rights, i.e. rightsconditional upon individual requisites that must be carefully and closely verified and monitored through time.Moreover, the low level of institutional autonomy of the administrative system in some parts of SouthernEurope has made it difficult for officers in charge of benefit delivery to stand up to external pressures. As aresult, the relationship between benefit administrators and beneficiaries has often come to be mediated by“brokerage” structures between local and central authorities, sometimes linked to political parties or thetrade unions (see Cazorla 1992, 1994; Ferrera 1996). To these factors, one must add the traditionally poorintegration of social assistance (administered by different authorities and subject to different rules), resultingin eligibility overlaps and gaps in coverage.

Since social assistance benefits are typically granted on the basis of the means test, the capacity ofadministrators to assess “need” with some degree of accuracy is an essential requirement. Poor capacity isdoomed to lead to poor performance, if not to outright failure. The combination of extended households,high rates of self-employment, large informal economies, tax evasion and an institutionally weakadministrative apparatus has created peculiarly infertile grounds for the establishment and consolidation ofsocial safety nets in Southern Europe. On such grounds, it is objectively very difficult for administrators tojudge the material circumstances of applicants and thus their “real” eligibility to benefits (Atkinson 1998).Especially in Italy and Greece, fears of triggering off or exacerbating “welfare patronage” have been oftenevoked as politicoinstitutional justifications for limiting the scope of targeted schemes, except for thosebased on relatively straightforward categorical criteria (such as old age or physical impairment).

Highly segmented labour market regimes, unbalanced systems of social insurance, patchy and ineffectivesafety nets: these were the main problematic features shared by the Southern European welfare states at theend of the long and “glorious phase” of post-war development. The 1980s and, more explicitly, the 1990shave, however, witnessed not only a growing awareness of these three problems, but also the launch of asequence of reforms aimed at gradually “recalibrating” institutional and spending priorities and atstrengthening institutional capacities for the governance of social policies (Ferrera and Hemerijck 2003). Asthe three chapters on Greece, Spain and Portugal clearly illustrate, the return of democracy played a majorrole in re-orienting the political agenda in these countries: building a modern welfare state, aligned with EUstand ards, became a national goal (Gullén et al. 2003; Guillén and Matsaganis 2000). In Italy, the wish togain admission into the European Monetary Union (EMU) by the established deadline of 1998 gave in itsturn a decisive spur to modernizing reforms (Ferrera and Gualmini 2004). Tackling the three structuralproblems mentioned above has not been an easy operation. But despite socio-economic and politicaldifficulties, a rather ambitious agenda has been pursued in the four countries since the early 1990s, centredon the following ingredients: (1) attenuation of the generous guarantees for historically privilegedoccupational groups (especially in the field of pensions), accompanied by an improvement of minimum or“social” benefits (the various categorical safety nets, in the plural); (2) the expansion and amelioration of

INTRODUCTION 7

family benefits and social services with explicit attention to gender equality and equity issues; (3) measuresagainst the black economy and tax evasion; and (4) the reform of labour market legislation with a view topromoting de-segmentation and modification of unemployment insurance benefits.

For the purposes of our book, the most interesting ingredient of the recalibration process of the 1990s inSouthern Europe has been the new interest in the issues of poverty and social exclusion, the inauguration ofnew policy initiatives in the field of social assistance and, in particular, the spread of minimum incomeschemes. The successful launch of the Revenu Minimum d’Insertion (RMI) in France in 1988, together withthe new EU discourse and strategy on “inclusion”, set in motion developments that led to the adoption ofsimilar schemes throughout Southern Europe. Variations of RMI were adopted in the Basque Country in1988, in Madrid and Catalonia in 1990 and in other Spanish regions later during the decade. A pilot schemewas introduced in Portugal in 1996 and became fully operational on a national scale in 1997, while in Italy aformal experiment was launched in 1998 and further extended in 2000 (although it was subsequently“frozen” in 2003). Only in Greece has this trend so far been resisted, but even in this country the issue ofstrengthening the safety net (in the singular, i.e. through a generalized means-tested, but universallyaccessible scheme) has gained visibility. The national policy trajectories since the early 1990s are put incontext and briefly reviewed in the next sections, starting with Greece and ending with Portugal, i.e.Southern Europe’s success story in the 1990s.13

Greece: big debates, but little progress

The early origins of the Greek welfare state date back to the 1920s. Compulsory social insurance against oldage, invalidity and sickness was introduced in 1928 for all wage earners. As Matsaganis illustrates at thebeginning Chapter 2, in the subsequent decades the Greek welfare state developed, following an erraticcourse, giving rise to a plethora of occupational schemes (often micro-schemes for tinyprofessional categories), managed in a clientelistic fashion and plagued by chronic under-funding.

The restoration of democracy in 1974 ushered in a period of welfare state expansion, accelerated after thesocialist landslide in the 1981 general election. The unprecedented growth in social spending was aresponse to the expectations nurtured by large sections of society over decades of politically motivateddiscrimination. Greece’s accession to the European Community in 1980, widely considered to be aguarantee of political stability, legitimized aspirations for levels of income and social protection comparableto those enjoyed by other Europeans (Guillén and Matsaganis 2000).

The rising level of social spending did not, however, significantly improve the performance of transferpayments in terms of poverty reduction. This apparent contradiction can be attributed to the nature of thecountry’s system of social protection. The Greek welfare state has traditionally placed great emphasis oncontributory benefits, and especially pensions, with little provision for non-insurable risks such as poverty.Contributory social insurance is perfectly suited to Fordist norms of long and uninterrupted careers. But, tobegin with, the Greek labour market has always fallen short of reaching even a weak version of nation-wide, mature Fordism. Second, employment trends in the last couple of decades have expanded the numbersof those unable to meet Fordist norms. The long-term unemployed, the young who have not yet worked,women with a patchy working history, individuals employed on temporary or part-time basis, illegalimmigrants, workers in the underground economy and others have thus become “social insurance outsiders”who lose out in welfare terms, often heavily.

Social assistance has remained not only marginal within public social protection, but has also beencharacterized by institutional and operational dysfunctions. Benefits are poorly integrated, administered asthey are by different agencies and subject to different rules. Their interaction leaves in place not a coherent

8 MAURIZIO FERRERA

whole, but an uneven structure that combines eligibility overlaps with coverage gaps. Given that non-contributory transfers are more naturally suited to the pursuit of anti-poverty objectives, the marginal natureof social assistance leaves a social safety net that in reality is full of holes, through which individuals andtheir families can slip into poverty. Poor households are ineligible for one of the existing benefits if they donot fit the “identikit” imagined by legislators, failing to fulfil the narrow categorical conditions required.

The safety net in old age is also patchy. Those with sufficient contributions are entitled to a minimumpension (since 1996) plus an income-tested supplement. Lower non-contributory pensions are paid tofarmers and to those with low income and no other pension entitlement. No universal minimum guarantee isavailable. Partly as a result, Greece features a unique combination of high spending on pensions and highpoverty in old age, at 13 per cent of GDP and 35 per cent of those over 65 respeo tively (Matsaganis 2002).

Fragmentation and incomplete coverage are evident in all other areas of social security. Unemploymentbenefit is contributory and of limited duration (12 months): on average less than half of the registeredunemployed actually claim benefits. Income transfers to families are targeted to those with three children ormore, so that poor children in smaller families receive little or no assistance. Disability benefits vary bycondition and recipient status (10 categories and 22 sub-categories of benefit). Housing assistance iscontributory and geared towards owner occupation, i.e. beyond the reach of the poor.

Overall, non-contributory benefits accounted for 16.3 per cent of all spending on social security in 2001,while income-tested benefits accounted for a mere 4.7 per cent. The gradual phasing-out of basic farmerspensions (the OGA scheme) since 1998 and the abolition of the income test on “many-children benefits” in2002 (9.3 per cent and 1.9 per cent respectively of all expenditure on social security) have further reducedthe space reserved for these two types of benefits within Greece’s social protection system.

However, selectivity has become a fashionable idea since 1996, when the socialist government under anew leadership declared EMU membership to be an overriding aim, while pledging its commitment to a“cohesive society”. Indeed, the concept of selectivity was hit upon as a rather obvious way to square thecircle. The strategy yielded some early results (the income-tested pension supplement mentioned earlier),but soon ran out of steam, presumably for lack of obvious targets in a social protection system stilldominated by contributory benefits.

The 2001 National Action Plan for Inclusion (NAP-incl.)—prepared in the context of the EU “socialinclusion process” mentioned above—implicitly ruled out the option of minimum income, while at the sametime reiterating a commitment to selectivity. A number of new measures, taking effect from 2002, wereannounced and subsequently implemented—with little success. The 2003 NAP-incl. adopted an even moreconservative approach, limiting itself to a strategy of marginal adjustments to fill the most visible gaps.

On the whole, the “danger that some groups experiencing poverty may not be eligible for incomesupport” (CEC 2001) remains largely undiminished despite some recent improvements. The absence of alast resort benefit, targeted in nature but universal in scope, remains a crucial missing link in the socialsafety net. Opposition to minimum income renders the anti-poverty armour of the social protection systemvulnerable. While the administrative difficulties involved in implementing such a scheme must not beunder-estimated, the financial requirement could be modest: as Matsaganis shows, a simulation exercise putthe cost of transfers under a minimum income programme in 2000 at � 269 million or 0.23 per cent of GDP.

Italy: grand reforms and patchy experiments

Of the four countries covered in this volume, Italy has the most mature welfare state. Compulsory socialinsurance was introduced between 1898 (work injuries) and 1919 (old age, invalidity and unemployment)for all employees. The 1945–75 period witnessed a sequence of expansive reforms, which created one of the

INTRODUCTION 9

most generous pension systems in Europe. In the first part of Chapter 3, Sacchi and Bastagli show,however, that poverty and social exclusion remained marginal issues in the national debate and policyagenda. Only since the mid-1990s have these issues gained political saliency and institutional attention. Astandard diagnosis of the historical failings of assistenza has emerged, widely shared by political actors andsocial partners, i.e. high fragmentation, policy overlaps, a bias towards transfers (against services), markedterritorial differentiation (Italy’s curse, even within Southern Europe), and the absence of a safety net of lastresort.

In 1977 responsibility for social assistance was devolved to regional and local tiers of government.Guiding principles and general standards were left for consideration by a national framework law thatwould regulate the whole sector but this law was not issued until 2000. Laws passed by various regionsallowed wide discretion at municipal level. Thus, the first minimum income schemes were the product ofmunicipal initiative. Turin in 1978, Ancona in 1981, Catania in 1983 and Milan in 1989 introduced a non-categorical means-tested benefit known as minimo vitale, even though many other large and importantmunicipalities (such as Bari or Rome) did not. Even in pioneering communes, the actual payment of thebenefit, however, depended on the availability of financial resources within local budgets.

At the national level, social assistance caters for specific categories such as the disabled and the elderly,but its role is marginal (6.4 per cent of total social expenditure in 2000). Social assistance measures includevarious non-contributory benefits, paying different amounts and subject to different types of means test.Moreover, low-income groups may be eligible for a number of social insurance benefits—such as familyallowances or pension supplements—which are graduated according to family size and income.

The concentration of resources on categorical and/or contributory benefits creates protection gaps at thebottom of the income scale. Things began to change in the mid-1990s, as various expert commissionsrecommended action to combat poverty and social exclusion. In particular, the Onofri Report (1997)provided a blueprint on which the subsequent reforms of social assistance were eventually based. The reportemphasized the need to enact national legislation on social assistance and the introduction of a generalized,non-categorical minimum income scheme. It also recommended phasing out some of the current socialassistance pro grammes and establishing a new mechanism to determine the financial situation of claimants.

Significant innovations followed. The Indicatore della Situazione Economica (ISE), a new set of rules toassess the material circumstances of potential claimants, was devised in 1998. ISE specifies how incomesand assets may be taken into account when assessing claims for means-tested benefits. The decision to takeinto account wealth as well as income mainly rested on practical considerations; it aimed to correct for theunfairness caused by tax evasion on the part of some categories of potential claimants (Baldini et al. 2002).ISE applies to two benefits introduced in 1998: Large Family Benefit (targeted to families with three ormore children) and Maternity Allowance (aimed at mothers ineligible for contributory maternity benefit).

Moreover, in 2000 a new framework law reformed the institutional setting of Italian social assistance,according to principles of decentralization and subsidiarity within national guidelines and performancestandards. This was an important and ambitious reform, which seriously tried to innovate on three delicatefronts: (1) universalism (through the promotion of policies addressed to all individuals and families in asituation of need and not only to certain categories); (2) provision in kind and services (rather than thetraditional cash transfers); and (3) reliable funding (through the establishment of a National Social Fund,regularly fed by the public budget on an annual basis). The reform of Title V of the Constitution in 2001,however, undermined the legal foundations of the framework law, while the advent of the Berlusconigovernment in the same year deprived it of the political steam necessary for serious implementation.

In 1998 a national experiment was launched in the field of minimum income. Partly modelled on theFrench experience, the Reddito Minimo di Inserimento (RMI) consists of a monetary and an “activation”

10 MAURIZIO FERRERA

component; entitlement to cash assistance is conditional on participation in insertion programmes. Cashassistance amounts to the difference between the income guarantee adjusted for family size and theresources available to beneficiaries, with some conditions. Most beneficiaries (93 per cent) lived in theSouth, where some municipalities devised ingenious ways to cope with the implications of the undergroundeconomy for income assessment. Overall, the first phase of the experiment (1998–2000) involved 34,700families in 39 municipalities. The budget law for 2001 extended the experiment for another two years andraised the number of eligible municipalities to 306. At the end of 2003, the Berlusconi government decidedto freeze the experiment, linking its future destiny to a pending new constitutional revision of centre-periphery relations.

The independent evaluation report on the first phase of the experiment (IRS et al. 2001) threw light onvarious aspects of the scheme. On the positive side, RMI—the missing pillar of Italian social assistance—breaks with a long tradition of categorical, discretionary income support to the poor that had provedineffective as an anti-poverty policy. Prior to 1998, residents in some participating municipalities wereineligible for any form of cash assistance even when in acute economic need. On this evidence, thegeneralization of RMI throughout Italy (foreseen, though not in an automatic way, by the 2000 frameworklaw mentioned above) may be considered the logical next step. The same report estimated the cost of RMI ata national scale from � 2.2 billion to � 3 billion (0.18 per cent to 0.24 per cent of GDP) in 2001—asignificant but not excessive figure compared to other welfare programmes.

Two serious obstacles seem to be blocking the way to a full implementation of the minimum incomescheme: (1) the weak institutional capabilities of local administrations; and (2) the specific socio-economicenvironment of the Italian South. As mentioned in the previous section, minimum income schemes are quitedemanding in terms of institutional capabilities and managerial skills. Moreover, there is a risk of functionaloverload. Rather than a programme of last resort, in southern municipalities the experimental RMI hastended to become “the only game in town”, i.e. the only source of legal income in areas plagued by chronicunemployment, where the vast majority of the resident population is made up of outsiders with no jobopportunities in the formal economy and no social insurance entitlements.

The 2001 NAP-incl. discussed the RMI experiment in positive terms and even indicated it as an exampleof “good practice”. But the new centre-right government, voted into office in 2001, seems disinclined toproceed to a generalization of RMI. It has so far left the matter to the regions, refraining from either settingnational standards or committing financial resources. The 2003 NAP-incl. is vague on the matter, eventhough a Pact signed between the government and the trade unions in the summer of 2002 mentioned thegeneralization of a means-tested “income of last resort”, in the context of a broad reform of unemploymentsubsidies. As Sacchi and Bastagli conclude, the institutional and financial future of the safety net in Italy islikely to remain uncertain for some time to come.

Spain: regional schemes, variable geometry

The process of welfare state building took off in Spain with the establishment of the National Institute forSocial Insurance in 1908 (Instituto Nacional de Previsión Social). After the Civil War (1936–39), compulsoryinsurance for wage earners was implemented in the 1940s; 1942 for sickness and 1947 for old age andinvalidity (the SOE and SOVI schemes respectively). Only during the desarrollismo (“developmentalism”)phase (1950s and 1960) did the Francoist regime take an interest in expanding social insurance—a trendthat culminated in the social security reform of 1972, which improved benefits against unemployment,disability and old age.

INTRODUCTION 11

Social assistance under the Franco dictatorship was meagre. The Church and the family were the mainproviders of welfare to the needy. The democratic Constitution of 1978, however, inaugurated a period ofinstitutionalization of social services and assistance. The Carta Magna left basic legislation and socialsecurity in the hands of central government, but social assistance became an “exclusive” competence of the17 Comunidades Autónomas (regions). As in Italy, these immediately claimed several functions withrespect to social assistance in their own constitutional charter (Moreno 2001). In 1982–93 regionalparliamentary acts established social service systems open to all citizens. As explained by Arriba andMoreno in Chapter 4, the upgrading of social services (and later the introduction of minimum incomeschemes) were used by the regions as an instrument to enhance popular legitimation and territorialcohesion, in an “imitation” process similar to what other authors have called “competitive state-building”(Banting 1995).

In 1987, an agreement between central, regional and local governments resulted in the approval of the“concerted plan for the development of basic provision of social services by the local authorities”, thatpromoted administrative co-operation between the three tiers of government. In 1988 a Ministry for SocialAffairs was created in Madrid, but the attempts by the PSOE to enact a National Law on Social Services (anante litteram counterpart of the Italian framework law of 2000, mentioned above) resulted in politicalfailure, owing to regional opposition.

Spanish entry into the EU gave additional impulse to policy innovation in the areas of poverty andexclusion, also in the wake of the country’s participation in the II European anti-poverty programme. In1988 (taking effect from 1990), old age and disability pensions were universalized, forming a fundamentalcomponent of the social safety net. Noncontributory benefits became available on a means-tested basis tothe elderly, the disabled and to low-income families with dependent children. Earlier, unemploymentassistance had been introduced in 1984 as a response to mounting joblessness.

In April 1995, a report on the “analysis of the structural problems of the social security system and of themain reforms required”, undersigned by all main political parties and trade unions, known as the “ToledoPact”, was ratified by the Congress of Deputies and became law. Its provisions included a clearer separationof contributory and non-contributory benefits, as a result of which universal health and social services andmeans-tested social assistance were to be fully financed through general taxation, while a reserve fund wasalso created within the contributory regime to strengthen its future viability. A number of other socialreforms have been introduced since the mid-1990s, geared towards the “activation” of claimants, theadoption of stricter criteria of access to unemployment assistance and the establishment of a personal andfamily minimum allowance in the form of a refundable tax credit.

Besides the improvement in social services, starting from the late 1980s, the regions began to implementminimum income programmes (Rentas Mínimas de Inserción), along the lines of the French RMI and in thewake of EU recommendations. The first such programme was introduced in the Basque Country in 1989.This move had a thoroughgoing demonstration effect and all the other regions followed suit in subsequentyears; the last scheme was introduced in the Balearic Islands in 1995. Approximately 80,000 families with200,000 members (0.5 per cent of population) benefited from these programmes in 2000 (see Table 1.2).

Regional programmes have a common purpose, but differ with respect to adequacy, coverage or means toachieve social integration of programme beneficiaries. Only the Basque scheme can be considered agenuine minimum income programme, based on a subjective and judiciable right. Well-developed schemesalso operate in Madrid and Catalonia. At the other extreme, some regions provide minimum incomeprogrammes of limited coverage at a low level—conditional upon resource availability in local budgets—ormerely offer temporary employment in “socially useful” projects (Aguilar et al. 1995).

12 MAURIZIO FERRERA

In 2000, the basic monthly rate (for beneficiaries living alone) varied from � 239 in the Canary Islands to� 305 in the Basque Country and � 319 in Extremadura and Navarre (for comparison, the minimum wageamounted to � 496 and non-contributory pensions at � 288 per month). Adjustments for family size are madeaccording to flat equivalence scales, while total benefit amounts are subject to a maximum limit. Totalexpenditure on minimum income benefits reached � 210 million (0.03 per cent

Table 1.2 Minimum income programmes (2000)

Greece Italy Portugal Spain

Income guarantee (�per month) , singleperson

148a 268 125 286e

Income guarantee,couple+2 children

444 660 374 386e

Number ofbeneficiaries(thousand)

700 86b 418 202

Number ofbeneficiaries (%population)

6.4 3.6b 4.2 0.5

Cost of minimumincome scheme (�million)

269 220c 284 210

Cost of minimumincome scheme (%GDP)

0.23 0.22d 0.25 0.03

Source: Chapters 2, 3, 4, 5.Notesa All figures listed for Greece, where no minimum income programme operates, are estimates from a simulation

exercise of the likely effects of such programme (reported in Matsaganis et al. 2001).b Number of beneficiaries (and percentage of local population) in the 39 municipalities participating in the first wave

of the minimum income experiment.c Cost of the scheme in the 39 municipalities over the 2-year period ending 31 December 2000.d Mid-point estimate of total cost in 2001 (% of GDP) if the scheme were generalized throughout Italy.e Value of minimum income benefit in Catalonia.

of GDP) in 2000. At the regional level, expenditure varied from � 337,000 in La Rioja (less than 0.1 per centof the regional budget), reaching � 53 million in the Basque Country (over 1 per cent of the regionalbudget). Catalonia, Andalusia and Madrid spent between them � 89 million (Arriba and Moreno 2002).

Future prospects for minimum income programmes are uncertain. Their implementation contributed tothe legitimacy of the new Comunidades Autónomas and was favoured by fiscal federalism: from 3 per centin 1981, regional spending accounted for as much as 33 per cent of all public expenditure in 2001. Regionshave been able to integrate social services and social assistance into common local networks of provision,but most programmes suffer from chronic underfunding. Moreover, their decentralized nature risksexacerbating regional disparities in welfare provision, though top-down harmonization can hardly beregarded as a viable option in a federalized country like Spain (Moreno 2003). Still, in a not-too-distant future,regions could face the dilemma of either requesting co-funding from central government or limiting thescope of existing programmes.

INTRODUCTION 13

As noted by Arriba and Moreno in Chapter 4, the 2001 and 2003 National Action Plans for Inclusionwere important occasions for intergovernmental co-ordination, providing the first synthetic overview of thefight against poverty in Spain. While there are no doubts about the institutional and financial solidity of theBasque or Catalan RMI, the prospects for the other regional schemes remain less firm and less certain. Asshown by Table 1.2, Spanish spending and coverage rates in this field are much lower than in Portugal—whichhas introduced a standardized national scheme—even if the two countries have comparable poverty levels: afigure that testifies to the incomplete maturity of many Spanish regional schemes. For some time to come,“variable geometry” is deemed to remain the most appropriate metaphor to capture Spain’s situation asregards minimum income.

Portugal: from pilot initiatives to a fully-fledged national RMI scheme

The first laws on social insurance were introduced relatively early in Portugal, in the same year as Italy:1919. In this country, however, compulsory insurance remained largely on paper and the task of providingsecurity to workers and their families continued to be performed by charitable institutions and friendlysocieties. The dictatorship of the Estado Novo did promote the development of corporatist insurance bodiesbut coverage remained gappy until the 1960s. Only in 1962 did a new law bring some order to the field andestablished state responsibility over social insurance schemes. As stressed by Capucha, Bomba, Fernandesand Matos in Chapter 5, during the long phase of conservative dictatorship, poverty was indeed widespread(affecting ca.40 per cent of the population), but was not a significant policy concern.

The democratic revolution of 25 April 1974 introduced—through the new Constitutional text—a set ofsocial rights and institutions formally instituting a modern welfare state. However, the adverse economicconditions in the ensuing period limited the financial and organizational resources needed to put the newlyestablished social policies into practice. In particular, measures to fight poverty (which still affected 30 percent of the Portuguese population in the early 1980s), remained scarce and social assistance was fragmented.

In 1984 social protection was thoroughly reorganized. A framework law defined the basis of socialsecurity as foreseen in the Constitution, setting out a three-level structure. The general regime providedcontributory benefits to workers and their dependants, while the non-contributory regime and socialassistance provided additional transfers and services not envisaged by the general regime. Anti-povertypolicies received a boost after Portugal’s entry into the EU in 1986, in the wake of Portuguese participationto the II European poverty programme that favoured actionresearch projects targeted on specific groups. Itsstress on participation made it compatible with the “local development approach” that had already beentried within the country (even during the dictatorship, through some pilot initiatives in the mid-1960s). Fromthen on, antipoverty policy relied on a territorial integrated approach and partnerships at national and locallevel (Almeida et al. 1994).

Between 1980 and 2000 Portugal witnessed an impressive increase in the level of social spending, whichgrew from 12.8 per cent to 22.7 per cent of GDP (more than halving its distance from the EU average, equalto 24.3 per cent in 1980 and 27.3 per cent in 2000). In the year 2000, social security legislation was revised;a new law aimed to raise benefits and to ensure the sustainability of social security by reinforcing the financialbasis of public pensions and by assigning full responsibility for social assistance to the national budget. Thereform was supported by an agreement with all social partners except the Industrial Employers Federation.The law built on a new generation of social policies, launched in 1995. These intended to “activate”individuals but also social welfare institutions, encouraging the latter to become more responsive through anindividualized approach to citizens’ needs and conditions

14 MAURIZIO FERRERA

The minimum income scheme Rendimento Mínimo Garantido (RMG) was undoubtedly the flagship ofthese new policies (Capucha 1998). In the run-up to the 1995 general election, a high-profile debate on EUCouncil Recommendation 92/441 on sufficient resources (issued under the Portuguese presidency of the EU)put poverty at the centre of the political agenda. The incoming socialist government introduced RMG on anexperimental basis in 1996 and extended it nation-wide on 1 July 1997.

The simplest description of the Portuguese RMG is as a contract; it pro vides income support in the formof a non-contributory benefit to guarantee a minimum living standard, in exchange for a commitment toparticipate in a social integration programme. The innovative character of RMG lies in that provision ofcash assistance is combined with participation in a range of activities involving integration into the labourmarket, return to education, vocational training, access to health services, housing interventions, etc. Theprogramme is open to all legal residents in situations of proven acute financial hardship. Beneficiaries mustbe at least 18 years, except for younger parents or pregnant women or carers of older people withdisabilities.

RMG offers a differential cash allowance, equal to the difference between the assessed net income andthe guaranteed amount for each household type. The latter is equal to the social pension for each of the firsttwo adults in the household, with increments of 70 per cent for any additional adult and 50 per cent perdependent child. The calculation of household resources excludes family allowances and student grants. Toease integration in the labour market, 20 per cent of earnings from work or training is also disregarded (50per cent in the first year). Since RMG is subsidiary to other social benefits, participants are required to claimfirst any benefit they may be eligible to. Benefit is initially awarded for a year, with the possibility ofautomatic renewal if beneficiaries continue to meet the requirements. Social workers from localadministration assess the economic situation of claimants. If substantial discrepancies between actual andreported incomes are discovered, there is a corresponding adjustment or even termination of the benefit.

The social worker in charge of each claimant produces an individual report, which must include adescription of the problems faced by the household and must propose an “integration plan”. This takes theform of an agreement signed by the social worker on behalf of the “local monitoring committee” and themembers of the household. The agreement specifies the tasks to be accomplished by beneficiaries and thesupport given by local institutions. The agreement may be renegotiated with the beneficiary if it isunsuitable or needs to be amended. Access to benefits is conditional on participation in the integration planand the whole process is legally binding.

By December 2001, 752,000 persons (7.5 per cent of the total population) had at some time participatedin the programme, of which 354,000 (3.6 per cent of the population) were still in receipt of benefit. Asignificant number of beneficiaries seem to have been reintegrated into society after a period of receivingminimum income assistance: of the 398,000 people who left the programme, 258,000 did so because they wereno longer in a situation of acute need. Total expenditure on minimum income benefits reached a peak of� 284 million (0.25 per cent of GDP) in 2000 and fell back to � 235 million (0.19 per cent of GDP) in 2001.Two difficulties have been identified in the aftermath of the 1997 law: the first concerns the ability tocontrol fraud, the other is activation for labour market participation (including the quality of insertion plansprovided by institutions). As Capucha and his colleagues argue, evidence shows, however, that frauds arerather insignificant, in spite of the hype surrounding the few cases actually detected. Work disincentives arealso relatively limited, since benefit values are low and activation measures act as an effective screeningdevice. On the other hand, the scheme faces a shortage of human resources, as the individualized approachand the need to monitor the complex process of reinsertion place a heavy burden on the limited number ofsocial workers.

INTRODUCTION 15

RMG once again became the subject of a political debate in the course of the 2002 election campaign.Although abolishing RMG was not on any political agenda, the need to cut public expenditure and the riskof a poverty trap were used to argue in favour of changes in the scheme. As a matter of fact, the centre-rightcoalition that had severely criticized the minimum income programme emerged victorious from thatcampaign. Significantly, however, once in office, the new centre-right government did not question theprinciple of a universal right to minimum income, but limited itself to introducing some minor changes tothe scheme. First, the scheme was renamed Social Insertion Income (RSI) with a view to underlining evenmore the “activation” component. Eligibility criteria have also been somewhat tightened and newmechanisms to limit fraud have been introduced. The Constitutional Court has judged some proposals totighten criteria as unconstitutional. On the whole, the essential features of the scheme's design are likely toremain unaltered for the time. The 2003-05 National Plan for Inclusion has confirmed the maintenance ofthe programme, with some fine-tuning. Future challenges could derive from budget constraints deprivingthe scheme of vital human resources.

Some comparative conclusions: Southern Europe as a model for transitioncountries?

As the preceding discussion illustrates, Southern European countries differ among themselves both in termsof the design of anti-poverty policies and of the institutional configuration in which such a policy operates.Yet they continue to form a distinct cluster as all four still face a similar set of challenges and have recentlyembraced a similar social policy agenda. In contrast to the institutional stagnation of previous decades, the1990s have definitely witnessed significant and promising policy innovations in the field of socialassistance. The new generation of targeted benefits and services, launched in all four countries, moved inthe appropriate direction; these new benefits and services filled some—often the most macroscopic—of thetraditional gaps in coverage, taking important steps towards the much needed recalibration of socialexpenditure vis-à-vis the risk of poverty and thus the most vulnerable groups in society.

Without doubt, EU initiatives have played a significant role in promoting such recalibration. Theinfluence of Council Recommendation 92/441 has already been noted. The launch of the social inclusionprocess in 2001 prompted in all four countries the first serious attempt at formulating a comprehensivediagnosis of current challenges and existing policies. With all their weaknesses (rightly highlighted by theauthors of the country chapters), the various National Action Plans for Inclusion do signal a genuine effort atfocusing policy attention on the dark spots of Southern European welfare and at outlining a more preciseagenda of reform. The increased attention on “social minima” and the very adoption of the safety netmetaphor have been encouraged by the EU discourse on cohesion, inclusion and guaranteeing sufficientresources—confirming the “maieutic potential” of EU soft law and of the OMC in particular (Ferrera et al.2002).14

The drive to establish effective minimum income guarantees has moved at different speeds and alongdifferent paths in the four countries. Portugal definitely stands out as the success story, as it was able tofully implement a national rights-based scheme. A number of factors have combined to pave the way forthis success and a first checklist would include at least three. The first factor was the existence of a domesticpolicy legacy (a “path”) of targeted anti-poverty initiatives at the local level, dating as far back as the1960s; a small, but significant institutional and “epistemic” capital was already there when Portugal joinedthe EC in 1986. The second factor was, precisely, the new set of spurs offered by the EU: the Poverty IIprogramme, which was under way when Portugal entered the EC and which was promptly joined by thePortuguese government; the large inflow of cohesion funds, whose overall “philosophy” fitted well with the

16 MAURIZIO FERRERA

domestic policy background (the first factor); the adoption of Council Recommendation 92/441 on theguarantee of sufficient resources, under the Portuguese presidency, which activated a “moral commitment”to implement the Recommendation objectives at home and at the same time established close links betweenthe Portuguese authorities and the European Commission (the DGV in particular). The third factor was thepolitical swing of 1995, which brought into office a relatively cohesive and politically ambitious centre-leftexecutive, interested in a broad strategy of social modernization and in the electoral dividends it couldgenerate. The RMG scheme immediately became a prime objective of this new strategy.

To some extent, similar factors were present in the other three countries as well but they were eitherweaker or their virtuous combination was inhibited by other counter-factors. In Spain, for example,accession to the EC and the presence of a socialist government, seriously committed to welfare statemodernization, did help in paving the way for the introduction of the various regional schemes of RentaMinima throughout the early 1990s. However, the decentralized nature of the Spanish polity and the failureof the central government, in 1988, to enact a national law of social assistance blocked the option of aPortuguese-style (or French-style) national scheme, or even the setting of common national standards. Asmentioned, a landscape of variable geometry has resulted in this country, which devotes to the regionalprogrammes of minimum income fewer financial resources than neighbouring Portugal, even though Spainis much bigger and displays similar levels of poverty.15

In Italy, regional decentralization (and, more generally, the territorial diversity and the backwardness ofthe Mezzogiorno) operated as an inhibiting factor in respect of the establishment of a national scheme or thesetting of national standards. At its onset, the 1998 pilot experiment promised to serve as the necessaryinstitutional wedge, capable of virtuously breaking with the tradition of regional improvizations and“chaotic” (rather than just variable) geometry. But the Constitutional reform of 2001, the arrival of the newBerlusconi government (less committed towards recalibration and unwilling to take to completion acontroversial policy measure of its predecessors) and, finally, the objective difficulties generated by the pilotscheme in the Southern context have rapidly eroded the innovation potential of the Reddito Minimod’Inserimento scheme.

In Greece—the country in which the least progress has been made so far —the problem has not beendecentralization, but rather the presence of a peculiarly “soft” public administration, a very stronginstitutional legacy of fragmentation and categorical programmes, and the enormous social and politicalresistance to any serious attempts at welfare state recalibration.

On the whole, the verdict on the new experience of the 1990s must remain open. On the one hand, thesigns of positive innovation are quite evident; the welfare states of Southern Europe have clearly embarkedupon an ambitious process of modernization, which includes the strengthening of their “bottom lines” in thesocial policy sphere. As outlined in the chapters of this volume, the systematic evaluation of the actualimpact of the new schemes in terms of poverty-reduction and “insertion” is still in its early stage. But, again,the first signs emerging from such evaluations do contain some “rays of hope”. On the other hand, however,Southern European safety nets still remain rather frail in terms of institutional design and—especially inGreece and Italy—also in terms of political support and legitimacy. The Italian case exemplifies howpromising reforms may soon prove to be false departures. Only in Portugal do the seeds planted in 1996–97seem to have undertaken a stable process of maturation—institutional, financial and, not least, political. Theresistance of the Portuguese minimum income scheme to the change in government in 2002 can beconsidered a good indicator of both institutional and political maturation —even though more time isneeded for full consolidation.

The weakness of the Southern European safety net should be a cause for concern; as the traditionalcapacity of Southern European families to respond to social needs declines, the pressure on formal social

INTRODUCTION 17

protection will inevitably intensify. Meeting this challenge will certainly require a sub stantial organizationaland administrative effort, though the fiscal cost of strengthening safety nets is likely to remain rathermodest.

Strengthening the social safety net also raises the question of the most appropriate level for action. This isespecially relevant in Spain and Italy, where the issue is intertwined, as mentioned, with the wider issue anddebates on subsidiarity and decentralization. While the presence and activism of sub-national levels ofgovernments (regions and municipalities) have undeniably opened up new opportunities for innovation andvirtuous experimentation at the regional level, they also raise the risk of perpetuating, even reinforcing,traditional gaps and disparities. On balance, the Spanish experience leans towards the former, whereasItaly’s emerging scenario points to the latter, especially in the wake of the 2001 Constitutional reform andof the Berlusconi government's inclination to shift responsibility of “last resort” forms of support to theregions.

While summarizing Italian development, we stated above that in some under-developed regions theexperimental RMI ended up becoming “the only game in town”, i.e. virtually the only legal source of incomefor vulnerable groups. This risk is certainly not exclusive to Italy: are there ways of contrasting this risk—inthe Mezzogiorno, but also in the predominantly rural areas of Spain, Greece and Portugal? The intensity ofthe risk is highly sensitive to the wider menu of social policy which is available in these areas. We canillustrate the point with the help of Figure 1.1.

Figure 1.1 The institutional location of RMI-type schemes in under-developed areas.

18 MAURIZIO FERRERA

Figure 1.1 aims to illustrate the proper institutional location of an RMI-type scheme in an under-developed socioeconomic context. This location is at the core of a configuration formed by at least threedistinct “squares”. First, there is an outer square consisting of (1) cohesion policies, aimed at ensuring aninflow of financial resources from richer areas, through either national or supranational arrangements ofterritorial redistribution; and (2) local development policies, aimed at mobilizing all the various external andinternal resources and “capitals” (social capital, institutional capital) towards the achievement of growthobjectives and, more importantly, towards the activation of local engines of growth (e.g. an industrialdistrict, specializing in a given production or service, including tourism). The second square consists in itsturn of two policy ingredients: (1) active labour market policies, aimed at fostering employment (inconnection with the local development project) and at supporting it with adequate placement services andtraining schemes; and (2) social services, offering to families—and especially women—support for mostacute caring needs. The third, inner square consists of the classical cash transfer schemes to protect theunemployed, on the one hand, and to compensate for family burdens, on the other (at least through targeted,but generalized, child benefits). In such a configuration, the RMI is located at the bottom (the safety netmetaphor) of the inner square in its cash component: it should only cater for the economic needs notadequately covered by the other policies and benefits which are in place. In its insertion component, thismeasure crosses over into the spheres of active labour market policies, on the one hand, and that of socialservices, on the other.

The current situation in much of Southern Europe is still far from this configuration. As mentionedabove, Southern European countries spend much less than the other EU countries on family benefits andservices. Though formally generous, unemployment benefits only reach a modest fraction of their targetpopulations, especially in Italy and Greece. Active labour market policies and social services are still inneed of development, especially in those areas where they are most needed. Cohesion policies have playedan important role as a source of income throughout the 1990s (in Italy and Greece, already during the1980s). But—again, with the partial exception of Portugal—they have not been systematically andeffectively linked to policies for local development.16 It should be no surprise that the risk of the RMIbecoming “the only game in town” has been so high, generating problems of functional overload. Theseproblems have been counteracted in Spain by de facto rationing the limited resources available in the poorerregions (where minimum benefits are actually paid only to the extent that local budgets allow it), in Italy byfreezing the pilot experiment and in Greece by not even moving from proposals to facts.

The future of anti-poverty policies and, in particular, minimum income programmes will ultimately reston political considerations. These pro grammes are vulnerable and so are their beneficiaries. Sustaining themomentum for reform depends on the ability and strength of transversal coalitions in their favour. At thismoment, it can only be hoped that the efforts of recent years to weave and mend social safety nets in theSouth will discontinue; social protection outsiders are still too many and still too poor.

To what extent are Southern European developments and recent innovations relevant to other regions ofthe world? This volume addresses the question in its last chapter, with reference to Eastern and in particularSouth-Eastern European countries. As Sotiropoulos illustrates in Chapter 6, the welfare state of the so-called“real socialism” was a “caring patronage state”, formally universalistic in its rules of coverage and inclusion,but de facto highly particularistic to the extent that personal and political ties (not to mention outrightcorruption) often filtered access to benefits and services. After the fall of the Berlin Wall, all post-socialistcountries have undertaken a difficult journey to more Western-type forms of social protection—a journeywhich has not only crossed troubled waters, but that has also generated (especially in the former SovietUnion republics) new and dramatic social problems. Partly owing to the influence of the World Bank, themetaphor of the safety net has gained increasing salience during the transitions and various categorical

INTRODUCTION 19

measures (often on an ad hoc basis) have been experimented with in most countries of the former Socialistbloc—with quite modest success.

While recognizing the big differences between Southern European and Eastern European experiences,Sotiropoulos highlights a number of features that make a comparison between the two worth pursuing—especially if one looks at the “South of the East”, i.e. countries such as Rumania, Bulgaria, Albania, Croatiaand Macedonia. Four elements stand out in particular in these countries: (1) the high incidence ofagriculture and the ensuing presence of a pronounced urban-rural divide; (2) a wide-spread informaleconomy and concomitant tax evasion, which produce a skewed (and highly opaque) income distributionand high rates of “false positives” in social assistance programmes (i.e. the inclusion of non-poor people inpoverty alleviation schemes); (3) the persistence of extended and traditional forms of households, with allthe ensuing implications in terms of gender; and (4) limited administrative capacity, both on the revenueside (low capacity to raise taxes) and on the spending side (low capacity to manage means-tested schemes).

Based on these elements, it is not difficult to predict that South Eastern countries are very likely toembark (if they have not already done so) upon a path to welfare state building which will bear manyresemblances to that followed in the past by Greece, Italy, Spain and Portugal (probably in that order). Thelabour market, social insurance and social assistance systems of these countries have high probabilities ofevolving towards the segmented, functionally and distributionally unbalanced conditions displayed bySouthern European countries between the 1970s and early 1980s—with the additional challenge of muchmore disadvantaged starting conditions and a much more turbulent international economic environment. Aspointed out by Sotiropoulos, symptoms of the insider-outsider cleavage, of relatively generous socialguarantees for the former and macroscopic gaps of coverage for the latter are already clearly visible inSouth-Eastern Europe.

Can Southern Europe “teach” anything to her poorer Eastern neighbours? Policy transfer is a trickyinstitutional operation and it seldom works. There are, however, weaker forms of cross-national policy linksthat might play a positive “bridging” role between the South-West and the South-East of Europe; let usthink of “benchmarking”, of initiatives for studying and taking inspiration from good practices (but alsofrom other countries' failures), of actions for promoting the formation of expert and more generallyepistemic communities that can facilitate and channel innovation from one national context to the next. Onthese fronts there is much that could be done or at least experimented with—including loose forms of “openco-ordination”. As recognized by Sotiropulos, the policy blueprint presented in Figure 1.1 might, forexample, provide useful insights and suggestions for the Eastern European systems—especially in terms ofthe range of policy buffers that ought to accompany an RMI scheme. Southern Europe is not used toconsidering itself as a social policy “model”, but, as the chapters of this book show, some promising lightshave started to glow in the South during the past 15 years. And besides, everything is relative, includingsocial policy models. For Eastern European reformers in search for ideas and inspiration, the Greek, Italianand Iberian experiences are the most obvious and probably the most promising point of departure.

Notes

1 In the UK, the term “safety net” had already been used in the wake of the Beveridge Report during the 1940s and1950s, especially in debates about National Assistance benefits (see below).

2 For a review of trends and debates, see in particular, Abrahamson (2003); Barnes et al. (2002); Begg and Berghman(2002); Gough (2001); Muffels et al. (2002) and Saraceno (2002).

3 For details, see Acknowledgements. A short summary of the main findings of the FIPOSC project is offered inMatsaganis et al. (2003).

20 MAURIZIO FERRERA

4 The state of the art of this literature can be found in Gunther et al. (1995) and Morlino (1998).5 See especially Ferrera (1996 and 2000); Ferrera and Hemerijck (2003); Guillén and Álvarez (2001); Kuhnle

(2000); Moreno (2003); Petmesidou (1996) and Rhodes (1997).6 Family benefits are not explicitly mentioned by Scharpf and Schmidt, but they did constitute a central ingredient

of the social security package introduced for workers in many European countries between 1945 and the 1970s. 7 As will be illustrated in Chapter 5 by Capucha et al., Portugal constitutes a partial exception, given its high

employment rates, also on the part of women.8 It must be noted that the earnings opportunities provided by the informal economy have paradoxically

contributed to the maintenance of low wage levels in the formal economy, and thus to the high numbers ofworking poor (especially in Portugal).

9 To be sure, informal activities in small and very small businesses are present also in some of the more developedareas of Southern regions, such as Veneto or Catatonia.

10 For a detailed discussion of the underground economy in Southern European countries, see Ahn and De la Rica(1997) and Carillo and Pugno (2002).

11 The source is a study undertaken by the Italian trade union CGIL, illustrated by the daily newspaper Il Sole-24ORE of 5 May 2004, p. 14.

12 For a discussion and some estimates of child labour in Southern Europe, see ILO (1996).13 The four sections below have been prepared with the close collaboration of the authors of the various national

chapters and especially of Manos Matsaganis, to whom I wish to express my warm gratitude.14 The notion of a “maieutic potential” of the OMC was originally suggested by Stefano Sacchi.15 It must be noted, however, that figures on spending are probably under-estimated in Spain.16 Moreover, these funds risk a financial contraction in the future, in the wake of the eastern EU enlargement (see Viesti

and Prota 2004).

References

Abrahamson, P.A. (2003) “Researching poverty and social exclusion in Europe”, Journal of European Social Policy, 13(3):281–307.

Aguilar, M., Laparra, M. and Gaviria, M. (1995) La caña y el pez: el salario social en las comunidades autónomas1989–1994, Madrid: Foessa.

Ahn, N. and De la Rica, S. (1997) “The underground economy in Spain: an alternative to unemployment?”, AppliedEconomics, 29 (6):733–743.

Alber, J. (1982) Von Armenhaus zum Wohlfahrtsstaat, Frankfurt: Campus.Almeida, J.F., Capucha, L., Costa, A.F., Machado, F.L., Nicolau, I. and Reis, E. (1994) Exclusão social, factores e tipos

de pobreza em Portugal, Oeiras: Celta Editora.Arriba, A. and Moreno, L. (2002) Poverty, Social Exclusion and “Safety Nets”, UPC Working Paper 02–10, Madrid

(www.iesam.csic.es/doctrab).Atkinson, A. (1998) Poverty in Europe, Oxford: Blackwell.Baldini, M., Bosi, P. and Toso, S. (2002) “Targeting welfare in Italy: old problems and perspectives on reform”, Fiscal

Studies, 23 (1):51–75.Banting, K. (1995) “The welfare state as statecraft: territorial politics and Canadian social policy”, in S.Leibfried and

P.Pierson (eds), European Social Policy between Fragmentation and Integration, Washington, DC, BrookingsInstitution, pp. 269–300.

Barnes, M., Heady, C., Middleton, S., Millar, J., Papadopoulos, F., Room, G. and Tsakloglou, P. (2002) Poverty andSocial Exclusion in Europe, Cheltenham: Edward Elgar.

Begg, I.B. and Berghman, J.B. (2002) “Introduction: EU Social (Exclusion) Policy revisited”, Journal of EuropeanSocial Policy, 12 (1):195–205.

Capucha, L. (1998) Rendimento Mínimo Garantido: Avaliação da fase experimental, Lisbon: CIES/MTS.

INTRODUCTION 21

Carillo, M.R. and Pugno, M. (2002) The Underground Economy and the Underdevelopment Trap, Discussion PaperNo. 1, Trento: Dipartimento di Economia, Università di Trento.

Cazorla, J. (1992) Del clientelismo tradicional al clientelismo de partido: Evolución y Características, WP 92/55,Barcelona: Institut de Ciencies Politiques I Socials.

Cazorla, J. (1994) El clientelismo de partido en España ante de la opinion púublica, WP 94/86, Barcelona: Institut deCiencies Politiques i Socials.

CEC (2001) Draft Joint Report on Social Inclusion: Communication from the Commission, COM (2001) 565 final,Luxembourg: Office for the Official Publications of the European Communities.

CEC (2002) The Social Situation in the European Union, Luxembourg: Office for Official Publications of the EuropeanCommunities.

Deacon, B. (2000) “Globalisation and social policy”, in UNRISD, The Next Step in Social Development, OccasionalPaper No. 5, Geneva: UNISRD.

De Swaan, A. (1987) In Care of the State, Cambridge: Cambridge University Press.Eardley, T. et al. (1996) Social Assistance in OECD Countries: Synthesis Report, London: HMSO.Fernández Cordón, J.A. (1997) “Youth residential independence and autonomy: a comparative study”, Journal of

Family Issues, 6:576–607.Ferrera, M. (1996) “The southern model of welfare in social Europe”, Journal of European Social Policy, 6 (1):17–37.Ferrera, M. (2000) “Restructuring the welfare state in Southern Europe”, in S. Khunle (ed.), Survival of the Welfare

State, London: Routledge, pp. 166–181.Ferrera, M. and Gualmini, E. (2004) Rescued by Europe? Italy’s Social Policy Reforms from Maastricht to Berlusconi,

Amsterdam: Amsterdam University Press.Ferrera, M. and Hemerijck, A. (2003) “Recalibrating European welfare regimes”, in J.Zeidin and D.Trubeck (eds),

Governing Work and Welfare in a New Economy`: European and American Experiments, Oxford: OxfordUniversity Press, pp. 88–128.

Ferrera, M., Matsaganis, M. and Sacchi, S. (2002) “Open co-ordination against poverty: the new EU ‘Social InclusionProcess’”, Journal of European Social Policy, 12 (3):227–239.

Gallie, D. and Paugam, S. (2002) Social Precarity and Social Integration. Report for the European Commission, Brussels:Eurobarometer 56.1.

Gough, I. (2001) “Social assistance regimes: a cluster”, Journal of European Social Policy, 2 (2):165–187.Guibentif, P. and Bouget, D. (1997) Minimum Income Policies in the European Union, Lisbon: Union dos Mutualidades

Portuguesas.Guillén, A. (2002) “The politics of universalisation: establishing national health services in Southern Europe”, West

European Politics, 25 (4):49–68.Guillén, A.M. and Álvarez, S. (2001) “Globalization and the Southern European welfare states”, in R.Sykes, B.Palier

and P.Prior (eds), Globalization and European Welfare States: Challenges and Change, London: Palgrave,pp. 103–126.

Guillén, A.M. and Matsaganis, M. (2000) “Testing the ‘social dumping’ hypothesis in southern Europe: welfare policiesin Greece and Spain during the last 20 years”, Journal of European Social Policy, 10 (2):120–145.

Guillén, A.M., Álvarez, S. and Adao e Silva, P. (2003) “Redesigning the Spanish and Portuguese welfare states: theimpact of accession into the European Union”, South European Politics and Society, 8 (1–2):75–98.

Gunther, R., Diamandouros, P.N. and Puhle, H.J. (eds) (1995) The Politics of Demo cratic Consolidation: SouthernEurope in Comparative Perspective, Baltimore, MD: The Johns Hopkins University Press.

International Labour Office (ILO) (1996) Child Labour: What Is to Be Done?, Document for discussion at the informalTripartite Meeting at the ministerial level, Geneva, 12 June.

IRS, Fondazione Zancan and CLES (2001) Valutazione della sperimentazione del Reddito Minimo di Inserimento,Rome: Presidenza del Consiglio dei Ministri.

Kuhnle, S. (ed.) (2000) Survival of the Welfare State, London: Routledge.Marlier, E. and Cohen-Solal, M. (2000) Social Benefits and their Redistributive Effect in the EU, Statistics in Focus

(Theme 3 9/2000), Luxembourg: EUROSTAT.

22 MAURIZIO FERRERA

Matsaganis, M. (2002) “Yet another piece of pension reform in Greece”, South European Society and Politics, 7 (3):109–122.

Matsaganis, M., Ferrera, M., Capucha, L. and Moreno, L. (2003) “Mending nets in the South: anti-poverty policies inGreece, Italy, Portugal and Spain”, Social Policy and Administration, 37 (6):639–655.

Matsaganis, M., Papadopoulos, F. and Tsakloglou, P. (2001) “Eliminating extreme poverty in Greece”, Journal ofIncome Distribution, 10 (1–2):40–57.

Moreno, L. (2000) “The Spanish development of Southern welfare”, in S.Khunle (ed.), Survival of the Welfare State,London: Routledge, pp. 146–165.

Moreno, L. (2001) The Federalization of Spain, London: Frank Cass.Moreno, L. (2002) “Mediterranean Welfare and ‘Superwomen’”, UPC Working Paper 02–02, Madrid

(www.iesam.csic.es/doctrab).Moreno, L. (2003) “Europeanisation, mesogovernments and safety nets”, European Journal of Political Research, 42

(2):271–285.Moreno, L. (2004) “Spain’s transition to New Risks: a farewell to ‘superwomen’”, in P.Taylor-Gooby (ed.), New Risks,

New Welfare: The Transformation of the European Welfare State, Oxford: Oxford University Press, pp. 137–160.Morlino, L. (1998) Democracy between Consolidation and Crisis: Parties, Groups, and Citizens in Southern Europe,

Oxford: Oxford University Press.Muffels, R., Tsakloglou, P. and Mayes, D. (2002) Social Exclusion in European Welfare States, Cheltenham: Edward

Elgar.Naldini, M. (2003) The Family in the Mediterranean Welfare State, London: Frank Cass.Onofri Report (1997) Commissione per l’analisi delle compatibilità macroeconomiche della spesa sociale (Relazione

Finale), Rome: Presidenza del Consiglio dei Ministri.Perez Diaz, V. and Rodriguez, J. (1994) “Inertial choices: Spanish human resources policies and practices’, Analistas

Socio-Politicos, Research Paper 2b, Madrid.Petmesidou, M. (1996) “Social protection in Southern Europe: trends and problems”, Journal of Area Studies, 9:

95–125.Rhodes, M. (ed.) (1997) Southern European Welfare States: Between Crisis and Reform, London: Frank Cass.Saraceno, C. (1994) “The ambivalent familialism of the Italian welfare state”, Social Politics, 1:60–82.Saraceno, C. (2002) Social Assistance Dynamics: National and Local Poverty Regimes, Bristol: Policy Press.Scharpf, F. and Schmidt, V. (eds) (2000) Welfare and Work in the Open Economy, 2 vols, Oxford: Oxford University

Press.Schneider, F. and Enste, D. (2000) “Shadow economies: sizes, causes, and cons equences”, Journal of Economic

Literature, 38 (1):77–114 . Trifiletti, R. (1999) “Southern European welfare regimes and the worsening position of women”, Journal of European

Social Policy, 9 (1):49–64.Viesti, G. and Prota, F. (2004) Le politiche regionali dell’Unione Europea, Bologne: II Mulino.Vivian, J. (1995) Social Safety Nets and Adjustment in Developing Countries, Occasional Paper No. 1, World Summit

for Social Development, March.World Bank (2000) Social Protection Sector Strategy: From Safety Net to Springboard, Washington, DC: World Bank.

INTRODUCTION 23

2Greece—fighting with hands tied behind the back

Anti-poverty policy without a minimum income

Manos Matsaganis

Introduction

Only a few years ago “the absence of a national income safety net” could be identified as a defining featureof the “rudimentary social assistance regime of south Europe”.1 It is rather characteristic of the pace ofchange in this area that what a few years ago held for all of Southern Europe now is only true for Greece.Indeed, as the other contributions to this volume explain, fully-fledged minimum income schemes nowoperate in Portugal and in certain Spanish regions, while an experiment involving a number of Italianmunicipalities is still in progress.2 Greece, on the other hand, seems so far unable to go beyond the level ofpolicy debates; as shown later on in the chapter, the option of introducing some form of minimum incomeguarantee is still considered too contentious.

But why does Greece stand out? Can perhaps the absence of a minimum income programme be explainedby reference to effective “functional equivalents” acting similarly in practice? Or, rather more prosaically,should this absence be simply considered a policy failure with farreaching consequences for theeffectiveness of the social safety net?

This chapter provides a background against which such questions can be assessed. More specifically, itbegins with a brief account of the factors that made the construction of a comprehensive social safety net amatter of low priority in the past. In particular, four points are discussed in some detail: (1) the role of thefamily; (2) the nature of unemployment; (3) the rural dimension of poverty; and (4) the challenge posed byimmigration. It then goes on to describe the institutional mismatch between current social protectionarrangements, on the one hand, and the requirements for an effective anti-poverty policy, on the other. The“pension bias” and the residual role of non-contributory benefits are singled out as the main characteristicsof a social protection system with a high fiscal cost and a low impact in terms of poverty reduction.

The analysis of poverty trends seems to confirm the hypothesis of a “perforated” social safety net thatfails to prevent the descent of a sizeable proportion of the population into extreme poverty. This is exploredmore systematically later on, where a rather extensive account of poverty alleviation policies is provided.The next section presents a preliminary evaluation of recent measures and concludes that, despite someprogress, these are unlikely to fill the gaps in the existing social safety net. Then the reasons why the issue ofguaranteed minimum incomes has failed to catch the imagination of policy-makers and the general publicalike are discussed, while a simulation of the effects such a programme might have in terms ofparticipation, fiscal cost and distributional impact is offered. As this analysis demonstrates, introducing aminimum income scheme could go a long way towards effectively mending the country’s social safety net,while it would also reduce relative poverty and eliminate extreme poverty at a modest cost. The chapter

concludes with a discussion of the political conditions for a coherent policy response to poverty and socialexclusion in Greece.

The development of poverty alleviation policies

A historical perspective

As is inevitable, the history of the welfare state is inextricably linked to the history of Greece as a moderncountry.3 For almost a century after the creation of the Kingdom of Greece (1830), social protection, as itwas not called at the time, was confined to government employees such as army officers and civil servants,as well as some high-risk groups of workers such as seamen and miners. In spite of a battery of sociallegislation introduced by the Liberals in the 1910s, a mere 10,000 workers were covered by social insurancein 1922 (Liakos 1996:98).

The Asia Minor Catastrophe of 1922 brought the period of state building and national expansion to atragic end.4 In its aftermath, up to 1.5 million refugees of Greek origin arrived in Greece proper, a country witha population at the time of about 5 million. A further disaster, “humanitarian” this time, was eventually averted,albeit with great effort and some international support. The first emergency relief programmes weresubsequently established with the aim of helping newcomers resettle and make a new start as best as theycould.

Social insurance legislation was passed in 1922–23 and again after the return of the Liberals ingovernment in 1928, with the active support of the International Labour Organisation.5 In the 1930s, thecreation of IKA (the “Social Insurance Foundation”) extended compulsory coverage to all private sectorwage earners. It provided earnings-related pensions for old age, invalidity pensions as well as medicalinsurance. IKA was originally conceived as a comprehensive scheme into which all pre-existing schemeswould be gradually incorporated. As the original legislative proposal followed a slow and tortuous path(from a draft bill presented by the Liberal government in 1932 to the revised text made into law by thePopular government in 1934, through to its eventual enforcement under authoritarian rule in 1937), much ofthe original radical content was neutralized.

By the end of the 1940s, a sequence of events from war to occupation and then civil war had left adevastated and divided country. Despite ambitious plans to the contrary, Greek social insurance continuedto follow the erratic course that was later to become familiar. On the one hand, fragmentation alongoccupational lines continued to be combined with preferential treatment reserved for the categoriesconsidered to be politically close to the post-war national order.6 On the other hand, the seemingly widescope of many legal provisions turned out to be significantly narrower in practice, through a combination ofhabitual infringement and chronic under-funding.

Unemployment insurance was finally introduced in 1954, with the creation of the “Organisation forEmployment and Unemployment Insurance”, renamed the “Manpower Employment Organisation” OAE�in 1969. At about the same time, the first steps were taken to provide systematic assistance in cash to somegroups thought to be particularly at risk. A national scheme for the blind was put in place in 1951, while ascheme for orphans and other “unprotected children” was established in 1960.

The creation of O� A (the “Agricultural Insurance Organisation”) in 1961 optimistically extended theprinciple of compulsory social insurance to farmers. Nevertheless, a familiar pattern of less-than-promisedfinancial support to the new scheme in practice reduced benefits to low old age and widows pensions, pluslimited medical coverage. Therefore, a gap was established, separating farmers from urban dwellers, that

GREECE: ANTI-POVERTY POLICY 25

has survived to this day. As a matter of fact, a contributory main pension scheme for farmers was onlyestablished in 1998, gradually to replace the non-contributory flat-rate pension still in place.

The military regime7 of 1967–74 was curiously active in the field of social assistance. In a bid to project apaternalistic image, a scheme for the protection of large families was introduced in 1972, providing cashbenefits for third and additional children. Moreover, legislative decree 57/1973 established a generalframework for the provision of social assistance to the “needy”, the handicapped and victims of naturaldisasters. In the same year, a new disability scheme (for the “deaf-mute”) was put in place.

The restoration of democracy and the extension of civil rights to all citizens ushered in a period ofwelfare state expansion, accelerated after the victory of the Socialists in the 1981 general election. Almostimmediately, the statutory minimum wage was raised by 40 per cent, leading to corresponding increases insocial insurance minimum pensions and all other benefits indexed to it. This was followed by a 100 per centincrease in the value of O� A pensions, eligibility for which was extended to women (whereas until thatpoint it had been confined to male heads of farming households aged over 65). Universalisation of coveragewas further pro moted with the creation of supplementary pension schemes to cover new groups and,significantly, the introduction in 1982 of the “pension for the uninsured elderly”, targeted to those aged over68 with insufficient contributions (below the minimum record required for access to a social insurancepension) and inadequate incomes.

Expansion often degenerated to excess. Contributory conditions became lax, nowhere more so than in thecivil service, where a mother could retire on a full pension after a mere 15 years of service. Access to earlyretirement on favourable terms on grounds of “hard and arduous occupations” was extended to reach one-half of employees insured with IKA. Invalidity pensions proliferated too (34 per cent of all new pensions in1984), as a result of local committees adopting a “broad view” if faced with applicants who lacked adequatecontributions but were quite healthy otherwise. Such distortions—combined with the results of a chronicfailure to reverse or slow down the fragmentation of the pension system— are still familiar features of thesocial policy landscape in Greece.

Social assistance also grew somewhat with the introduction of new disability benefits to cover additionalcategories of claimants, including a general scheme for the severely disabled. In the area of social services,a key development was the establishment of a nationwide network of open care centres, aimed to reduce oldpeople’s social isolation and dependence. On a larger scale, the creation of a “National Health System” in1983 was a decisive, if still incomplete, step towards the goal of a universal health service.

The unprecedented expansion of social spending in the 1980s can be better understood against thishistorical background. The expansionary measures described above were a response to the expectationsnurtured by large sections of the population in decades of politically motivated discrimination. Greece’saccession to the European Community in 1980, widely considered to be a guarantee of political stability,appeared to legitimise aspirations for levels of income and social protection comparable to those enjoyed byother Europeans.

The brief interregnum of Conservative rule (1990–93) was chiefly marked by a serious attempt at pensionreform. The final outcome of successive interventions was higher contributions and lower replacement rates,as well as a universal retirement age of 65 for both men and women —but only for those entering the labourmarket after 1993. “Acquired rights” were largely left untouched, except for a modest rise in contributionsand a gradual reduction of retirement age differentials to take effect from 2007. On another level, a littlenoticed provision of law 2084/1992 lowered the age requirement for access to the social pension (for thosewith incomplete contributory record) from 68 to 65. During the same period, widespread fears that the fallin fertility threatened the survival of the nation led to the introduction of new (higher) family benefitstargeted to families with three or more children. On the other hand, in a bid to soften the effect of actual or

26 MANOS MATSAGANIS

planned privatisations, the duration of unemployment benefit was extended from six to nine and then totwelve months.

The Socialists returned to power in 1993 announcing, among other things, a new pension reform to“correct” the one approved by parliament a year earlier. The long season inaugurated in 1996, when a“modernising” leadership under Prime Minister C.Simitis took office with pension reform high on theiragenda, came to the conclusion six years later when law 3029/2002 was approved by parliament.8

The 2002 legislation, though successful in temporarily defusing the issue, can only be described as timidand ineffective if judged against the magnitude of the problem as originally diagnosed in terms ofunsustainable finances and serious inequalities. The replacement rate was set at 70 per cent, as in IKA, upfrom 60 per cent for those who entered the labour market after 1 January 1993 and down from 80 per centfor older cohorts of civil servants and workers in nationalised industries and banks. Reference earnings wereextended to the last five years in the “special funds” and to the best five of the last ten years for IKAworkers. The minimum pension for those who started work since 1993 was fixed to 70 per cent of theminimum wage. A lower rate of minimum pension was introduced for those with at least 11 years and 8months of contributions (but less than the 15 years required for a full minimum pension). Otherwise, IKA wasrefashioned as the “single fund of wage earners” and was guaranteed state funding to the tune of 1 per centof GDP a year. The law included no provisions for a change in retirement age. In fact, as a result ofseniority pensions being made available to public sector workers, it is expected that 2.5 million workers willretire earlier, another 1.0 million will be unaffected, while no-one will retire later. On the whole, the 2002law seemed unable to make a significant difference to the financial position of the pension system, while itsegalitarian drive was disappointingly weak.

Otherwise, following the change of leadership in 1996, the government declared EMU membership anoverriding aim, but pledged its commitment to a cohesive society. The concept of selectivity was hit uponas the obvious way to square the circle. The new orthodoxy yielded some early results. The income-testedpension supplement EKA� introduced in 1996 had enabled the government to escape an earlier pledge torestore the link of minimum pensions to the minimum wage. The eligibility criteria for the new benefitincluded a test of incomes based on tax reports, a feature hailed as a significant innovation. The emphasis ontargeting manifested itself further in income restrictions on benefits paid to large families. The new strategysoon ran out of steam, presumably for lack of obvious targets in a social protection system dominated bycontributory benefits. In a rather curious reversal, income testing for access to the so-called “many-childrenbenefits” was abolished in 2002.

The gradual phasing-in of a new contributory main pension by O� A mentioned earlier is set to raisefarmers’ pensions to a level comparable to that of other social insurance schemes. Meanwhile, thesignificant improvement in the value of all flat-rate pensions (i.e. including the social pension) by 75 percent in real terms since 1994 has gone some way towards bridging this gap in the shorter term.

Over the same period, significant developments in the rather neglected area of personal social servicesalso took place: a “home help” programme was successfully piloted and then further extended, while aNational Social Care Organisation was created to render more coherent the disparate initiatives taken in thisfield. Moreover, health reform is being slowly implemented providing, among other things, for a limited co-ordination of social health insurance, professional hospital management and a family doctor scheme. In thearea of social assistance, the proliferation of separate schemes for various types of disability continued(currently there are twenty-three different benefit rates), while the number of beneficiaries has increased bya factor of 4.4 since the mid-1980s.

The National Action Plan 2001–03 confirmed the commitment to selectivity and, among various otherinitiatives, announced three new benefit schemes. At the same time, it implicitly ruled out the option of

GREECE: ANTI-POVERTY POLICY 27

implementing (or even piloting) a guaranteed minimum income scheme. This stance was reiterated andmade more explicit in the National Action Plan 2003–05. Such official opposition to minimum incomesrenders the antipoverty armour of the social protection system vulnerable and makes Greece a deviant caseif set against the policy innovations taking place over recent years in every other Southern Europeancountry. This issue is discussed in more detail later in the chapter.

The legacy of the past

As the preceding section hinted, poverty in Greece seemed for a very long time to be at the same time amatter of general concern and of low political priority. The paradox can probably be explained by theperception of poverty as a common problem, rather than as one that affects only some groups in society.When Greece became part of the EC/EU and found itself to be its poorest member, the drive to “catch upwith Europe” was widely assumed to be of utmost urgency and, incidentally, an effective anti-povertypolicy. There is no doubt that such a version of the “trickledown theory” was responsible for the relativeneglect of poverty as a policy issue. Still, that is only part of the truth. Poverty would not have remainedsuch a marginal issue for such a long time were it not for the simple fact that it only rarely led to what isnow called “social exclusion”. Four points appear to stand out: (1) the role of the family; (2) the ruraldimension of poverty; (3) the nature of unemployment; and (4) the challenge posed by immigration. Thesepoints are briefly explored, and their relevance to poverty brought out, in the following paragraphs.9

The role of the family

Historically the family in Greece has functioned as an effective (albeit informal) social safety net. It pooledresources to help the young buy a house or start up a business, it provided unpaid female carers to look afterchildren and elderly relatives, it acted as a lending institution in case of emergency, it even supplied jobs,either in the family business or through kinship networks. For all the distortions some of these practiceshave caused, the strength of family ties has contributed to a relatively cohesive society. It can be argued thatthe significance of family-provided welfare (and the “privatisation” of many social problems which thisentailed) have made the need to strengthen social safety nets seem less pressing than might have been thecase otherwise.

In recent times, average household size came down as the evolution from extended to nuclear family wasalmost completed, to the extent that three-generation households have now become a rarity. Nonetheless,despite the rapid demographic, social and cultural changes of the last decades, there is evidence to suggestthat family ties and kinship networks have not been substantially weakened, but simply adjusted to the newconditions.10 The low, by European standards, frequency of divorce or births out of wedlock points to thesame direction.

Obviously, excessive reliance on the family as a provider of welfare has a very serious side effect: thatthose who, for various reasons, are excluded from family solidarity may have few social resources left todraw on. At the same time, the family itself is under stress and can hardly be expected to perform acomparably supportive role in the future. The spectre of smaller families having to look after an increasingnumber of older relatives, and often being unable or unwilling to do so, is a powerful warning and a starkreminder of future challenges to social policy in Greece.

28 MANOS MATSAGANIS

The rural dimension of poverty

Social cohesion in Greece was reinforced by the predominantly rural, until recently, character of thecountry. The proportion of population employed in agriculture and/or living in rural areas has certainlydeclined in recent times. Nevertheless, the relevant figure still remains higher than in most other Europeancountries. The rural dimension is doubly significant, as poverty rates are markedly high in rural areas,especially among retired farmers. Yet, old people living in rural villages are a classic case of a social groupthat experiences poverty but only rarely social exclusion, as most of them are perfectly well integrated intotheir local communities.

The forward march of urbanisation poses a threat to traditional patterns of social cohesion. Remote anddepopulated villages that are almost exclusively inhabited by the elderly are hardly in a position to shareresources or practise much solidarity. At the same time, informal support networks are difficult to replicatein city neighbourhoods or out-of-town suburbs where the overwhelming majority of families now live. Thegradual transformation from a mainly rural to a mainly urbanised society raises difficult issues for socialpolicy—not all of which have been successfully resolved yet.

The nature of unemployment

Unemployment in Greece has risen in recent years to reach 12 per cent in the late 1990s, compared to a EUaverage of 8 per cent. Currently, it stands at just below 11 per cent. However, contrary to what has been thecase elsewhere in Europe, the rise in unemployment was not accompanied by a similar rise in poverty,which remained quite stable.

Close examination of the incidence of joblessness explains why. Unemployment among “prime age”males is relatively low, while the young and women (especially, young ones) bear the brunt of highunemployment. The gender gap is compounded by the fact that female labour participation is below 50 percent of the economically active population aged 15–64, compared to approximately 60 per cent for the EUas a whole. The combined effect of unemployment and non-participation is shown by the employment rate:according to data from the 1999 Labour Force Survey, as many as 92.4 per cent of men aged 30–44 worked,while the corresponding figure for women of the same age was only 56.8 per cent.

The implication of the above is that unemployment typically—and to a greater extent than elsewhere inEurope—threatens not so much the “male breadwinner” as his wife and their children. In other words, the maineffect of unemployment on the social fabric is not poverty for entire families. Rather, it is “frozenfamilialism”, as the survival (on life support) of the male breadwinner model at the expense of women andthe young might be termed.

The challenge of immigration

In a society historically characterised by strong family ties and wellintegrated rural communities, a strongtradition of solidarity developed. However, that was a particular type of solidarity, often limited to membersof family or kinship networks, from which “outsiders” could be excluded. The contradictions of such a(paradoxically exclusive) notion of solidarity were recently exposed by the large presence of foreignimmigrants. Until the early 1990s, Greece was a very homogeneous society from the point of view of ethnicmake-up as well. At about this time, the descent of the Balkans into severe economic crisis, civil unrest oroutright war led to an estimated 600,000 foreign immigrants settling in Greece. In a country of only 10million, this made for a very considerable proportion of the population. Indeed, while immigration has

GREECE: ANTI-POVERTY POLICY 29

become a permanent feature of the social landscape throughout Europe, nowhere else has the transition fromemigration to immigration been completed at such speed as in Greece.11

The social tensions arising from such massive population movements made progress towards integrationslower and more uneven than might have been otherwise. Indeed, legislation permitting foreign workers toapply for a work permit (a first step towards full legalisation) was only approved in 1998. As a result ofthat, those foreign workers who are registered enjoy full social insurance rights.12 Nevertheless, socialprotection for the many unregistered immigrants is limited to emergency services provided either byhumanitarian organisations or by the informal support networks of the immigrants’ own communities.

Summing up: “old” poverty rarely precluded social integration, while “new” poverty is primarily aproblem that affects outsiders. These two related reasons may be part of the explanation for the low policypriority given to poverty so far.

Inevitably, the historical “invisibility” of poverty as a distinct policy issue is reflected in the marginal roleof anti-poverty policy within the social protection system as a whole. This theme is developed below.

A case of institutional mismatch

Social protection in Greece, as elsewhere in southern Europe, is commonly held to lag behind compared tothe rest of Europe. This perception can be shown to be inaccurate. As an examination of the evidencereveals, in quantitative terms Greece has converged with the European Union as a whole to a considerabledegree. More specifically, revised estimates13 show that social expenditure in Greece, at least as aproportion of GDP, has risen steadily over recent years (Table 2.1).

While social spending statistics seem to suggest that Greece has in recent years been catching up with therest of Europe, some caution in interpreting these figures is necessary. Indeed, it should be remembered thatthe aggregate level of expenditure on various social protection programmes is no more than an indicator of acountry’s “welfare effort”. Although important, examination of total expenditure alone is unhelpful

Table 2.1 Total expenditure on social protection (1990–99)

1990 1993 1996 1999

Greece 22.9 22.0 22.9 25.5

EU-15 25.5 28.8 28.5 27.6

Sources: ESSPROS data reported in Abramovici (2002): 2. Data of the “European System of Integrated SocialProtection Statistics” (ESSPROS) include explicit as well as implicit expenditure, on benefits in kind as wellas in cash, incurred not only by government or social insurance agencies but by employers or other privateentities as well, provided the relevant benefits are “collectively available”.

if the issue of interest is the level of social protection enjoyed by the beneficiaries “on the ground”. In thatrespect, a more complete analysis should assess the country’s welfare effort in terms of what might betermed the “social effectiveness” of social protection programmes.

The concept of social effectiveness is not easily defined, as different social programmes have differentaims, or multiple ones at the same time. For instance, health services aim to fight illness and promotehealth, employment services aim to fight unemployment and promote employment, social services aim toprovide assistance to various groups and help them integrate better in society, etc. The same goes for socialbenefits. For example, pension schemes attempt to promote saving and guarantee an adequate income atretirement, sickness or maternity allowances provide temporary income replacement, social assistance

30 MANOS MATSAGANIS

benefits aim to ensure that sufficient resources are available to those who might otherwise be at risk ofextreme poverty, and so on.

In view of the above, it is quite inappropriate to reduce the various aims of social protection programmesto a single measure. Nevertheless, it cannot be denied that fighting poverty is, or ought to be, one of themost fundamental aims of a country’s social protection system as a whole. On this count, Greece fares veryunfavourably in comparison to almost every other European country. The impact of social transfers inEurope has been analysed on the basis of European Community Household Panel data, a common survey offamily incomes in EU countries from 1993 onwards.

The data show that, in terms of poverty reduction, the impact of social transfers remains much weaker inGreece than in almost all other EU countries. The picture that clearly emerges regarding the effectiveness ofsocial transfers is disconcerting. In the absence of social transfers (notwithstanding counterfactual effects),poverty rates would have obviously been higher than they really are in all countries. In relative terms,however, “original income” poverty would have been lower in Greece than in the EU as a whole. On thecontrary, when the effect of social transfers is taken into account, the relative position of Greece vis-à-visEurope in terms of poverty is reversed. As a glance at the figures reveals, the Greek handicap in terms ofanti-poverty performance can be fully accounted for by the effect of social benefits other than pensions(Table 2.2).

This apparent contradiction (a high and rising level of social spending, on the one hand, combined with aweak effectiveness of social transfers, on the other) can be attributed to the nature of the country’s systemof social protection. Social insurance is suited to “Fordist” norms of long and uninterrupted careers. By thesame token, the long-term unemployed, the young who have not yet worked, women with a patchy workinghistory, individuals employed on temporary or part-time basis, illegal immigrants, workers in the shadoweconomy and others become “social insurance outsiders” who lose out often heavily in welfare terms.

Table 2.2 Distributional impact of social transfers (1999)

Relative poverty on the basis of:

original income before transfers original income+ old age pensions final income after all transfers

Greece 38 22 21

EU-15 40 24 15

Source: European Community Household Panel data presented in Dennis and Guio (2003): 6.

As a matter of fact, the welfare state in Greece places great emphasis on contributory benefits, with littleprovision for non-insurable social risks such as poverty, while social services remain at a very early stage ofdevelopment. As a consequence, while pensions account for the greatest part of social transfers, policiesaimed at families with children, the disabled, the unemployed and other groups at risk of poverty are far lessdeveloped. Indeed, in spite of certain recent developments, the “pension bias” remains one of thefundamental characteristics of the welfare state in Greece. This bias is clearly reflected in the compositionof social security (defined as the aggregate of social transfers in cash, whether contributory or not, incometested or not), as shown in Table 2.3.

In contrast, social assistance has remained a “poor relative” within the welfare system. Those benefitsthat do exist are poorly integrated, administered as they are by different agencies and subject to differentrules. Their interaction leaves in place not a coherent whole, but an uneven structure that combineseligibility overlaps with coverage gaps. Given that non-contributory transfers are more naturally suited tothe pursuit of anti-poverty objectives, the marginal nature of social assistance leaves a social

GREECE: ANTI-POVERTY POLICY 31

Table 2.3 Expenditure on social security benefits (2001)

� million % of total % of GDP

Retirement benefits 16,843 90.3 12.90

Family allowances 523 2.8 0.40

Unemployment benefits 514 2.8 0.39

Sickness benefits 427 2.3 0.33

Disability allowances 293 1.6 0.22

Housing benefits 44 0.2 0.03

Total benefits 18,644 100.0 14.28

Of which: non-contributory 3,041 16.3 2.33

Of which: income-tested 877 4.7 0.67

Source: Own estimates on the basis of data collected from benefit agencies and other sources.

safety net that in reality is full of holes, through which individuals and their families can and do slip intopoverty.

In fact, poor households are ineligible for any of the existing social benefits when they do not fit the“identikit” imagined by legislators, failing as a result to fulfil the narrow categorical conditions required bythe various programmes. Examples include:

• the long-term unemployed whose eligibility to benefit has been exhausted;• new entrants to the labour market who do not yet qualify for unemployment benefit because they have

never been employed;• those precariously employed with no social insurance entitlements to draw upon in the event of

temporary loss of earnings;• those retiring on a low pension (e.g. a social or survivor’s pension) and few other resources;• people incapable of working who for various reasons have failed to establish a claim to an invalidity

pension or one of the disability benefits;• low income families with less than three children, i.e. too few to qualify for any family allowance worth

claiming.

The list could easily be extended.As the above discussion implies, policy analysis predicts a weak impact on poverty of the social benefits

system overall and a higher risk of poverty among those groups that are exposed to gaps in coverage. Theformer was briefly touched upon in this section, but is revisited in the next—where the question of whetheror not the latter is brought out by available evidence is also examined in some detail.

The anatomy of poverty

No official definition of the poverty line has been adopted in Greece, nor does the term “poverty” figureoften in official documents. The National Action Plans for Social Inclusion are a rare but partial exception,as they devote considerable space to poverty estimates, even though their main aim is to show that suchestimates are misleading. Below we attempt to provide a snapshot of poverty in Greece drawing on allavailable sources.

32 MANOS MATSAGANIS

Relative poverty

EUROSTAT estimates show that relative poverty rates in Greece are among the highest in the EU. Morespecifically, according to the most recent ECHP data, the proportion of households with an equivalentincome below 60 per cent of the median was 20 per cent in Greece, compared to 15 per cent in the 15member states as a whole.14

The same data show that relative poverty in Greece is deeper (9 per cent vs 5 per cent in the EU-15below the 40 per cent of median line) and more persistent (13 per cent vs 9 per cent below the 60 per cent ofmedian line for three consecutive years) than in most other EU countries.

Alternatively, studies using Household Budget Survey data estimate the poverty rate at approximately 17–18 per cent (see Table 2.4). The difference is accounted for by imputed rents from owner-occupied housingand self-consumption of agricultural produce. These two features are clearly relevant to the task of correctlyidentifying the population in poverty. It seems reasonable to assume that not taking into account imputedincome overstates the gap that separates Greece from other European countries. However, the extent of theerror is unknown, given that Household Budget Survey (HBS) data on Greece and ECHP data on the EU-15cannot be compared without similar adjustments in other countries.

In any case, certainly the most remarkable fact about poverty in Greece is its stability over time. In fact,the relative poverty rate appears hardly to have changed at all since the early 1980s (Table 2.4).

A similar picture of stability is suggested by the trend followed by the poverty gap over the same period.Again, the proportion of total disposable income that is needed to lift all poor households to exactly the levelof the poverty line appears to have fluctuated around the 2.5 per cent mark from the mid-1980s to the late1990s (see Table 2.5).

The breakdown of poverty by household type, employment status, residence area, schooling record, etc.provides useful insights to the analysis. Relevant indicators are:

Table 2.4 Relative poverty rate (1981–99)

1981–82 1987–88 1993–94 1998–99

Income 17.26 17.01 18.44 17.25

Expenditure 17.63 18.70 17.13 17.88

Sources: HBS data analysed in Tsakloglou and Mitrakos (forthcoming). The poverty line was set at 60 per cent of themedian of each year’s distribution. The OECD modified equivalence scale used assigns values of 1.0 for thefirst adult, 0.5 for additional adults and 0.3 for children.

Table 2.5 Relative poverty gap (1981–99)

1981–82 1987–88 1993–94 1998–99

Income n.a. 2.43 2.46 2.40

Expenditure n.a. 2.64 2.16 2.24

Sources: Tsakloglou and Mitrakos (forthcoming). The poverty gap is the aggregate shortfall from a 60 per cent ofmedian poverty line as proportion of total income or expenditure.

• the poverty rate (i.e. the proportion of group members with equivalent income or expenditure below thepoverty line);

• the relative incidence of poverty (i.e. the poverty rate of the group in relation to the poverty rate of thewhole population);

• the group’s contribution to aggregate poverty (i.e. the proportion of all poor who belong to the group).

GREECE: ANTI-POVERTY POLICY 33

Interpretation of estimates requires some caution, as it is perfectly possible for a group to experience a highpoverty rate in relation to the general population and at the same time to contribute little to aggregatepoverty because its population share was small to start with. This seems to be the case of single parenthouseholds who face higher than average poverty rates (at least, in terms of income), though only a smallproportion of the poor are members of such households. The opposite is true in the case of families with oneor two children who have a low relative incidence of poverty but still account for a sizeable proportion ofthe poor population. This is shown in Table 2.6.

As the data imply, among the (partly overlapping) groups experiencing the highest rate of poverty areelderly couples, large families, households in which the “breadwinner” is unemployed, farmers or residentsof rural areas and households with low education. In this sense, a familiar picture of “old” poverty emerges,arguably destined to lose much of its importance as education standards improve, families shrink in size,people move to cities, the generations who lived through war and disruption gradually

Table 2.6 Relative poverty profiles (1999)

Population group Relative poverty rate Relative incidence Contribution to aggregate

Elderly couple (at least one 65+) 30.3 1.76 18.8

Single parent household 22.6 1.31 2.0

Couple with 3+ children (all <18) 29.5 1.70 8.2

Couple with at least one child 18+ 14.7 0.85 27.5

Couple with 2 children (both <18) 11.9 0.69 11.2

Unemployed 33.1 1.91 7.1

Private sector employee 15.8 0.91 19.0

Farmer 28.2 1.63 15.8

Self-employed (not in farming) 19.7 1.14 9.2

Pensioner 22.0 1.27 33.7

Resident in Athens 9.6 0.56 19.2

Resident in rural areas 32.5 1.88 41.0

<9 years of schooling 25.2 1.46 70.8

12 years of schooling 7.7 0.45 18.4

University or equivalent 4.1 0.23 3.5

Total 17.3 1.00 100.0

Sources: Tsakloglou and Mitrakos (forthcoming) and Appendix to NAPincl Greece (2001): 12.

pass away, etc. However, the poverty experience of specific groups can only provide a partial profile. A keyquestion concerns the relative weight of such groups in terms of their population share. When that is takeninto account as well, as in the last column of Table 2.6, the poverty profile looks quite different. On thebasis of their contribution to aggregate poverty, further groups show up in significant numbers: householdsheaded by private sector employees or the urban self-employed (between them, more than one-quarter of allpoor) and families with grown-up children (again, more than one-quarter of all poor). Even the “standard”type of couple with two children—whose low relative incidence of poverty is often noted—accounts formore than one-tenth of all poor. The significance and policy implications of this point are discussed lateron.

34 MANOS MATSAGANIS

Absolute poverty

The concept of absolute poverty recalls a “basket of goods” approach. In this, absolute poverty is defined asan income below the amount required in order to purchase goods or services considered essential for a sociallyacceptable living standard. An alternative course would be to fix the relative poverty line in real terms andmonitor over time the proportion of population falling below it. It is the latter that is followed here, with thepoverty line fixed in real terms at 60 per cent of the median of the 1974 distribution (see Table 2.7).

As the data suggest, absolute poverty varies quite considerably according to whether the incomedistribution of each Household Budget Survey is adjusted on the basis of National Accounts data or not. Onthe whole, there is a downward trend, though less uniform than might have been expected.

Extreme poverty

Social safety nets at best aim to eliminate extreme rather than broadly defined poverty. In other words, theguaranteed minimum cannot

Table 2.7 Absolute poverty (1981–99)

1981–82 1987–88 1993–94 1998–99

I. Unadjusted

Income 7.19 7.89 6.42 4.35

Expenditure 4.81 5.33 4.88 3.13

II. Adjusted

Income 8.25 7.37 6.84 8.83

Expenditure 6.48 6.29 3.12 2.33

Sources: The poverty line was held constant in real terms throughout the period at 60 per cent of the median of the1974 distribution. The poverty rate in 1974 was 20.29 per cent (income) or 20.39 per cent (expenditure). SeeTsakloglou and Mitrakos (forthcoming).

realistically be fixed at half average or 60 per cent of median income, the poverty line conventionally usedin statistics and academic research. Fiscal constraints, labour supply effects and, not least, politicalconsiderations dictate that the minimum guarantee be set at a lower level.

Extreme poverty might be thought of either in terms of a low proportion (e.g. 40 per cent) of medianincome, or as a living standard below the minimum implicitly—if not explicitly—set within a country’ssocial protection system. For analytical purposes, extreme poverty in Greece was defined by reference to thelevel of the pension for the non-insured elderly in 2000. The poverty line was adjusted for household sizeand composition.15 The resulting extreme poverty profile of Greece’s population in terms of householdcomposition, employment status and schooling record of household heads is shown in Table 2.8.

The risk of extreme poverty was found to decline very steeply with education, as the overwhelmingmajority of those below the threshold had a record of no more than nine years of schooling. In terms oflabour market attachment, extreme poverty was highest among households headed by an unemployedperson, though (given the differences in population share) pensioner and farmer households appeared tocontribute to aggregate extreme poverty much more significantly. With respect to composition of household,the social group experiencing the highest rate of extreme poverty was elderly people living with theirspouses, while other house

GREECE: ANTI-POVERTY POLICY 35

Table 2.8 Extreme poverty profiles (2000)

Population group Extreme poverty rate Relative incidence Contribution to aggregate

Elderly couple (at least one65+)

14.2 2.58 25.6

Single parent household 5.3 0.96 5.0

Couple with 3+ children (all<18)

4.8 0.87 1.0

Couple with at least onechild 18+

4.6 0.84 21.7

Couple with 2 children(both <18)

3.7 0.67 12.2

Unemployed 12.9 2.35 9.2

Farmer 12.2 2.22 24.4

Pensioner 11.1 2.02 33.5

Private sector employee 3.5 0.64 14.9

Self-employed (not infarming)

2.8 0.51 12.4

<9 years of schooling 8.9 1.62 88.4

12 years of schooling 2.1 0.38 9.8

University or equivalent 0.4 0.07 1.4

Total 5.5 1.00 100.0

Sources: Matsaganis et al. (2001). The extent and composition of extreme poverty were estimated using the EuropeanCommunity Household Panel database (1995 wave, uprated to 2000). The poverty line for a single individualwas set equal to the value of social pension (approximately � 1,800 a year in 2000). The equivalence scaleused assigns values of 1.0 for each of the first two adults, 0.7 for each additional adult and 0.5 for each child.

hold types clustered around the average. However, a considerable proportion of all persons in extremepoverty lived in families with older children or with two young children.

The case of two-child families is illustrative of the plight of population groups who fail to attract theattention of social policy-makers, on the grounds that they appear to be doing rather well in terms ofaverage living standards. Partly as a result of that, family benefits are mostly reserved for families withthree children or more (see next section). As a consequence, most children below the poverty line live inhouseholds that do not qualify for any family benefit worth claiming. By way of illustration, private employeesmay qualify (depending on contributory record) for a monthly family allowance of � 5.87 or � 17.61 (in 2003),for one or two children respectively.

In fact, the income composition of households in extreme poverty seems to confirm the above point.Roughly 40 per cent of all income of such households came from pensions, 28 per cent from self-employment and 19 per cent from wages, while other social benefits accounted for only 6 per cent of theirincome.

Housing conditions

At this point it is important to note that—almost by design—household budget surveys and similar statisticsfail to “capture” the experience of two groups that are believed to suffer particularly high poverty rates: the

36 MANOS MATSAGANIS

homeless (and others with no fixed address) and foreign immigrants, especially if illegal. In both cases,housing appears to be a key dimension to the poverty experience of these “statistically excluded” groups.

In fact, a recent survey of the housing conditions of low-income households in the area of greaterAthens16 found that foreign immigrants, compared to the general population of households in the survey,were more likely to do the following:

• rent rather than own (91.8 per cent, vs 26.1 per cent);• share with other family units (14.7 per cent vs 1.9 per cent co-habit);• experience crowded conditions (17.5 vs 27.5 sq.m. per person);• have sub-standard accommodation (11.2 per cent vs 3.9 per cent lack basic amenities).

Finally, non-governmental organisations estimate the extent of homelessness at 0.29 per cent. The figureconcerning Greek nationals only is 0.08 per cent, while the remainder is accounted for by immigrants orrefugees.17

The preceding analysis seems to confirm the hypothesis of a “perforated” social safety net failing toprevent the descent into extreme poverty of a considerable share of the country’s population. Furthermore,the characteristics of households below the threshold seem closely to correspond to the gaps in socialprotection identified earlier. These are explored more systematically in the next sections.

Poverty alleviation policies

This section offers a quite detailed presentation of the main policies targeted at low-income households ordeemed to have an impact on poverty.18 Each policy is classified under one of the headings ofunemployment, retirement, disability, family, housing and emergency benefits. More recent policieslaunched in the context of the social inclusion process are reviewed in a later section.

Unemployment benefits

Unemployment insurance is mandatory for all employees except tenured civil servants and agriculturalworkers. The “Manpower Employment Organisation” (OAE� ), a tripartite organisation financed byemployee and employer contributions and state subsidies, runs a variety of unemployment compensationschemes.

The most important of these is “ordinary unemployment benefit”. Eligibility rules require a contributoryrecord of at least eighty insurance days per year in the past two years before claiming (provided that 125insurance days were in the past fourteen months not counting the last two months). Alternatively, 200insurance days in the past two years not counting the past two months can also be sufficient. There is noincome test. Students, first-time job seekers and (implicitly) the long-term unemployed are not eligible.Benefit duration is five to twelve months depending on contributory record. Average duration is 7.08 months.Benefit is paid monthly, including a holiday bonus when appropriate.

The base benefit amount stood at � 264.39 monthly in 2002. On top of the base amount, allowances fordependent members, contribution bonuses, etc. may also be paid. The benefit rate was originally set attwothirds of the minimum wage “if resources permit”. However, the relevant ratio in 2002 was only 48.2 percent. Due to the strict contributory requirements, coverage is rather limited. The number of recipients was259,000 or 54.8 per cent of all registered unemployed.

GREECE: ANTI-POVERTY POLICY 37

A variety of “extraordinary unemployment benefits”, with less stringent rules, are also available. Theseinclude a five-month benefit for first-time job seekers aged 20–29 who can prove they have been out ofwork for over a year, lump-sum assistance for former recipients still unable to find work after eligibility tounemployment benefit has expired, special schemes for seasonal workers and irregular workers, etc. In1997, the latest year for which data were available, there were about 125,000 recipients of “extraordinaryunemployment benefits” as a whole.

Retirement benefits

The safety net in old age is patchy. Those with sufficient contributions may be entitled to a minimumpension plus an income-tested supplement. Lower non-contributory pensions are paid to farmers and to thenon-insured. No universal minimum guarantee is available. Partly as a result, Greece has a uniquecombination of high spending on pensions (13 per cent of GDP) and widespread poverty among the elderly(35 per cent of those over 65). The principal schemes are described in some detail in Box 2.1.

BOX 2.1RETIREMENT BENEFITS RECEIVED BY LOW-INCOME GROUPS

The social pension (or “pension to uninsured elderly”) is the one social assistance benefit in Greece thatcorresponds to the definition of a last resort benefit, albeit one targeted at the elderly. The benefit is funded outof general taxation and distributed by O� A (the “Agricultural Insurance Organisation”). Eligibility is reservedfor people over 65 years of age who are not in receipt of a contributory pension from a social insurance scheme.The age limit had been 68 years until 1993. The benefit rate is set equal to the flat-rate non-contributory oldage basic pension also provided by the same organisation. Receipt of another O� A basic pension by a spousedoes not cause disqualification, provided the couple has no additional income. Benefit rules specify that socialassistance benefits are disregarded in the assessment of applicants’ income. In the past ten years, there has beenan almost twofold increase in real terms. The value of the social pension for a pensioner without dependants in2002 was � 156.13 a month, paid fourteen times a year. Small supplements of � 2.60 a month per dependant alsoapplied. The number of recipients was 40,000.

The pensioners’ social solidarity supplement (EKA� ) is funded out of general tax revenue but distributed bythe social insurance organisations (pension funds) from which the recipient receives a pension. Pensioners ofO� A are excluded on the grounds that their pension is not contributory. EKA� was introduced in 1996, originallyaimed at low-income recipients of old age and survivor pensions over 65 (or of invalidity pensions irrespectiveof age). The limit was lowered in 1997 to cover old age pensioners and survivors aged 60–64. The benefit is paidfourteen times a year. Eligibility is determined on the basis of an income test in which three types of incomeare considered: (1) personal net income from retirement benefits and employment earnings; (2) personalincome from all sources; and (3) total family income. In 2002 the full rate (� 96.51 a month) was paid toclaimants with annual incomes below (1) � 5,474; (2) � 7,012; and (3) � 10,912, respectively by type of income.Reduced rates are paid to those with type (1) incomes up to 10 per cent above the threshold. Benefit rates andincome thresholds are revised annually by ministerial decree. Approximately 365,000 pensioners claimedEKA� .

The basic pension to farmers (O� A) is a non-contributory benefit awarded to men (and, since 1982, towomen) aged 65 and over, not in receipt of other pension, who have been active for at least twentyfive years inagriculture or similar sectors (such as fishing) and living in rural areas. Approximately 620,000 people receivedan old age basic pension from O� A. For about 45 per cent of them this was the only pension they received,though another 45 per cent had access to a low supplementary pension. The rest claimed the new contributorypension introduced in 1998, though still on a pro rata basis. The value of the basic pension for farmers is

38 MANOS MATSAGANIS

exactly the same as the pension for non-insured elderly, that is � 156.13 a month in 2002 (raised to � 170.80 in2003), paid fourteen times a year.

Minimum pensions are paid to approximately 320,000 old age pensioners of IKA (the “Social InsuranceFoundation”). A large number of survivor pensioners also benefited. The 2002 level of a minimum old age IKApension was � 377 monthly, paid fourteen times a year. About 35 per cent of that amounted to an implicitsupplement added to the “organic amount” (i.e. the part corresponding to contributions paid). Implicitsupplements to minimum pensions cost IKA approximately 20 per cent of the organisation’s total pensionexpenditure. Similar instruments exist in other social insurance organisations such as OAEE (the fund for theself-employed).

Invalidity pensions, notwithstanding the caveats mentioned earlier, at least potentially represent an importantanti-poverty instrument. The total number of invalidity pension recipients is approximately 350,000. Invaliditypensions are paid fourteen times a year.

Disability benefits

Disability benefits are available on a non-contributory basis, funded out of general taxation, butadministered by local authorities at the prefecture level .

Various forms of means testing were applied in the past with respect to most disability benefits. Theprefectures’ social service departments did carry out means tests of sorts, but practices differed. In 1987 aministerial decree attempted to standardise rules by fixing income thresholds that varied with family status,household size and age of beneficiary, but were otherwise identical across type of disability. A new meanstest procedure was introduced: income was to be self-reported through a signed statement, supported byprevious year’s income tax report. However, fourteen months later the decree was revoked afterprotestations of officials about the impracticalities and disproportionate costs of controls. A new ministerialdecree was issued in February 1989, abolishing all explicit income testing of disability benefits. Althoughno longer explicitly income tested, most disability benefits adhere to a logic of “subsidiarity”, as they arereduced or withdrawn altogether if the recipient is either in employment, or an old age pensioner, or inreceipt of invalidity pension from a social insurance organisation.19

As mentioned earlier, disability benefits are highly heterogeneous by type of disability and (sometimes)by category of recipient. More specifically, there are ten categories of disability with twenty-twosubcategories, in addition to a generic “mobility allowance”.20 More than 55 per cent of beneficiaries claimsevere physical disability benefit. Benefit levels are not indexed, but fixed by ministerial decree. Since themid-1990s the real value of disability benefits has risen by almost 70 per cent. The average monthlypayment in 2002 was � 183.35. Payments usually take place every two months, six times a year. The numberof benefit recipients in the same year approached 140,000 (up from 80,000 in 1994).

The Ministry of Health and Welfare, formally responsible for disability benefits, reported that thedistribution of funds earmarked for disability benefits between prefectures did not reflect “real need”.Moreover, it expressed fears that these funds were being partly spent by prefectures on other budget items.On the other hand, it noted the rise in number of beneficiaries of severe physical disability benefit inparticular (from 30,260 in 1994 to 73,150 in 2002) and proposed tighter application of procedures on thepart of awarding committees to contain it. As a result of the above factors, it was reported that many largeprefectures were unable to pay out benefits in the last two months of 2000.

GREECE: ANTI-POVERTY POLICY 39

Family benefits

“Many-children benefits” (), comprising 3rd child benefit, large family benefit and lifetime pension tomothers of many children, were all introduced in 1992. The benefits are funded out of general taxation, butdelivered by the Agricultural Insurance Organisation O� A. No contribution conditions apply.

In 1997 “many-children benefits” benefits became income-tested. Since then, benefit rates have beenregularly ungraded on the basis of the consumer price index, but income thresholds determining eligibilitywere only adjusted once, in 2000. In a little noticed policy reversal, income requirements for access to“many-children benefits” were dropped in 2002.

Other family benefits include contributory allowances and minor schemes such as unprotected childbenefit and birth grants to uninsured mothers. Contributory family allowances reinforce the large familybias of family benefits, as the amount of assistance increases almost exponentially with the number ofchildren (a pattern evident in the structure of tax reductions for children as well). Moreover, familyallowances introduce a new cleavage between civil servants, on the one hand, and most private sectorworkers, on the other, as the former receive much more substantial and timely assistance than the latter.Since many poor children live in families with one or two children not headed by civil servants, thestructure of family assistance leaves them exposed to the risk of poverty.22 The main schemes are describedin Box 2.2.

BOX 2.2FAMILY BENEFITS RECEIVED BY LOW-INCOME GROUPS

Third child benefit was originally paid to all families, irrespective of income, with a third child aged lessthan 3. Several significant changes were introduced in 1997: the benefit was raised for the first time (by about18 per cent), while eligibility was extended with respect to the child’s age (from 3 to 6 years of age), as well asbeing restricted with respect to annual family income. Before it was abolished in 2002, the threshold wasapproximately � 23,500 a year. The monthly rate in 2002 was � 141.03 and the number of recipients was 38,342.

Large family benefit is targeted at families with four or more children. The level of benefit is a function of thenumber of children who are aged less than 23 years and unmarried. In 1997, when the income test wasintroduced, the benefit amount per eligible child was raised by 16 per cent, while a minimum rate was set forlarge families with less than three eligible children. However, the minimum rate has been left unchanged innominal terms since then. When the income test last applied, families with an annual income in excess of � 29,350 (increased by 10 per cent for each child after the fourth) were ineligible for the benefit. In 2002 thenumber of receiving families was 82,008, with an average of 4.12 eligible children per benefit unit. Theaverage monthly payment per family was � 149.51.

A pension to mother of many children is paid to mothers of four or more children who are no longer eligiblefor “large family benefit” on the grounds that all the children are grown up. Before 2002 the income thresholdwas � 10,270 a year. The number of recipients in 2002 was 183,584, while the monthly rate was � 81.09.

Unprotected child benefit is a non-contributory benefit, delivered by local government at the prefecture levelunder the auspices of the Ministry of Health and Welfare. Beneficiaries are low-income single parent families,or households who “protect” orphans to whom they are related (Le. foster families are not eligible). The annualincome threshold, last adjusted in 1997, was set at � 2,820 for a three-member family, increased by � 250 foreach additional member. In 2002 the sum of � 44 per month was paid to 34,440 recipients.21

Birth grant to uninsured mothers is a lump-sum birth grant, paid to mothers who are ineligible for socialinsurance maternity benefits, whether as workers or as dependent family members, by virtue of being

40 MANOS MATSAGANIS

uninsured. The amount of � 440 (unchanged since 1997) was paid to each of the estimated 800 mothers whoclaimed the benefit in 2002.

Moreover, contributory family allowances are paid to civil servants and, through the “ManpowerEmployment Organisation” (OAE� ), to insured private sector employees irrespective of social insuranceaffiliation (see Table 2.9). Separate arrangements may also operate in some sectors, for instance, in banking, asa result of collective agreements.

Until 1998, OAE� paid lower rates of family allowance to high-income families, but a flat rate has been paidsince 1999. Claimants must have a contributions record (at least fifty days in the previous year, unless in receiptof unemployment benefit or incapable of working because of illness or disability). Payments are made once ayear. It is believed that take-up is low, though no estimates are available. No such restrictions apply withrespect to civil servants’ family allowance, which is paid automatically as a supplement to the basic salary on amonthly basis.

Table 2.9Contributory family allowances (2002)

No. of children OAE� Civil servants

1 70 211

2 211 423

3 475 845

4 578 1,409

5 674 2,289

NoteFigures are annual benefit rates in euros.

Housing benefits

Housing assistance is mostly contributory and geared towards owner occupation, therefore beyond the reachof most low-income families. The main schemes are described in Box 2.3.

BOX 2.3HOUSING BENEFITS RECEIVED BY LOW-INCOME GROUPS

A rent subsidy is paid by the OEK (the “Workers Housing Organisation”). Like OAE� , the Organisation isfunded through payroll contributions and covers all private sector employees, irrespective of social insurancemembership. Rather complex contribution requirements apply: single applicants must have a contributionsrecord of at least 3,000 days, but the requirement is reduced to only 1,000 days for families with five childrenor more, to 900 days for single mothers, and to 500 days of contributions if severely disabled. Young couples,temporary workers, Greek-origin return migrants, residents of remote areas and others are similarly favouredby lower contribution requirements. Homeowners and beneficiaries of other OEK social housing programmesare ineligible. The rent subsidy is income-tested. More generous income requirements apply for the elderly andyoung couples. Income thresholds increase with the number of dependent children. For example, the annualincome threshold for a young couple with two children was � 12,150 in 2002. Benefit rates also increase withthe number of dependent children (� 125 a month for a young couple with two children in 2002). When the

GREECE: ANTI-POVERTY POLICY 41

applicant’s family income exceeds the specified threshold, the benefit amount is first reduced pari passu andthen withdrawn completely if it falls below one-third of the full rate. A further restriction concerning maximumamount of rent allowed was abolished in 2003. The number of subsidised households was 36,768.

Moreover, prefectures distribute a housing benefit to uninsured elderly, funded out of general taxation.Applicants are required to submit a signed statement that they (1) live in rented accommodation and own nohouse; (2) are unable to meet their housing costs as they receive no social insurance pension and have noadditional income; and (3) are over 65 years of age. The benefit is paid directly to landlords on the basis of acontract with the local prefecture’s social services department. In 2002 the benefit rate for a single claimantwas � 116.20 monthly, raised to � 154 for a couple. The benefit was paid to 810 households.

Emergency benefits

A variety of other schemes pay benefits to return migrants, former detainees and victims of natural disasterson a discretionary basis. These are all funded out of general taxation and handed out by prefectures, exceptfor benefits to former detainees and unemployed return migrants administered by OAE� . All benefits are paidas lump sums. Benefits to return migrants were worth � 117.39 (first aid), � 176.08 (relocation expenses),� 293.47 (funeral expenses) and � 264.39 (unemployment). Benefits to victims of natural disasters were setat � 586.94 (first aid), � 4,402.05 (serious injury) and � 5,869.41 (damaged house). Former detainee benefitwas � 158.63. Finally, an emergency benefit for persons in need because of sickness or death of family headpaid � 234.78. All rates refer to lump sum payments in 2002. No reliable data on recipient numbers and totalexpenditure are available.

Redistributive impact

It was reported earlier that the anti-poverty impact of social benefits in Greece is much weaker than it is inthe EU as a whole. In addition, a recent study23 found that poverty reduction in Greece is mostly achievedby retirement pensions. By comparison, the impact of sickness and invalidity benefits is small, that offamily benefits negligible, whereas unemployment and “other” benefits seem almost to have no impact atall. The statistical appendices to the National Action Plans for Social Inclusion tell a similar story, thoughthey use a different source (the national Household Budget Survey rather than the European CommunityHousehold Panel). Compared to all other benefits put together, pensions have by far the largest impact onpoverty. The distributional performance of benefits seems to have improved somewhat in recent years, a factattributed to the increases to the basic pension for farmers and to the expansion of EKA� .

This section has presented the main policies already well established when the social inclusion process24

was set in motion. The approach followed in the two National Action Plans submitted so far and the othermeasures launched in their context are reviewed next.

The strategy against poverty and social exclusion

The National Action Plan for Social Inclusion, 2001–03

The first National Action Plan ruled out the option of moving towards a minimum income scheme in favourof a targeted approach, on the grounds that “the causes of poverty are complex and often vary amongdifferent vulnerable groups”. The general policy objective was identified as “to fill gaps in the social safety

42 MANOS MATSAGANIS

net”,25 to be mainly achieved through initiatives in the fields of social care and employment promotion(funded under the Community Support Framework).

In terms of social care, the “Home Help” programme has clearly been the most promising of recentinitiatives. The programme, started in 1998, targets frail elderly people living alone on a low income. Its aimis to allow participants to keep living in their own homes rather than in institutions, by supporting them witha variety of services. Local health visitors, psychiatric nurses, counsellors, as well as helpers withhousekeeping tasks such as shopping, cleaning, cooking and the like undertake to visit participants at homeon a regular basis. The programme recently expanded geographically and in scope (taking in youngerpersons with disabilities), and was strengthened through capital grants for the purchase of vehicles. “HomeHelp” has proved extremely popular with “clients”, while it has also helped boost employmentopportunities in social care.

In terms of income support, three new schemes (unemployment assistance for workers over 45 andrefundable tax credits for families with schoolchildren aged 6–16 and for those in less favoured areas) tookeffect from 2002. Two further schemes (social insurance contributions rebate and part-time work benefit)aimed at low earning workers had begun shortly before the 2001 National Action Plan for Social Inclusionwas submitted. All five schemes are described in Box 2.4.

The National Action Plan for Social Inclusion, 2003–05

In the second National Action Plan, the section dedicated to Common Objective 1.2 (“Facilitating access toresources, rights, goods and services for all”) opened with the statement that “guaranteeing minimumresources is a fundamental social objective”. It then went on to outline the two competing strategies forachieving that objective.26 The first strategy was the current policy of selective provision:

first, assistance is directed and support is provided where real need is identified and a mechanismexists for verifying and meeting that need; new benefits to cover other pockets of need are designedand implemented together with the infrastructure for delivery.

This was juxtaposed with “the immediate introduction of a generalised guaranteed income, thereby whoeverdeclares income below x is eligible for a benefit”. There were no prizes for guessing which of the twostrategies the authors favoured. “It is a basic principle of the NAPincl that such an approach in the presentconditions would harm the interests of needy citizens and defame social solidarity.” In such a way, “thegreat losers from a prematurely generalised blind guarantee would be those in real need. A blow will bedealt to the idea of an effective welfare state.”

As a matter of fact, the picture it presented of a steadily improving

BOX 2.4ANTI-POVERTY MEASURES INTRODUCED IN 2000±02

Unemployment assistance for older workers. As seen earlier, unemployment insurance implicitly excludesthe long-term unemployed (since eligibility expires after twelve months at the latest), while no generalunemployment assistance scheme exists. This new scheme aimed partly to fill this gap, being targeted at long-term unemployed workers aged 45–65 with annual family income below � 3,000 (plus � 587 per dependentchild). The benefit rate was fixed at � 150 monthly for a period of 12 months. The scheme, administered byOAE� , suffered from very poor implementation (see below) and made virtually no impact in the first two years

GREECE: ANTI-POVERTY POLICY 43

from its introduction. In 2003 the income threshold for eligibility was raised to � 5,000 per year and themonthly benefit rate to � 200.

Refundable tax credit for families with children at school aged 6–16. This scheme is aimed at families withannual income of less than � 3,000 with children aged 6 to 16. The benefit was conditional on school attendanceand amounted to � 300 per child per year. It was expected that the number of potential beneficiaries would be135,000 families at a cost of � 51.65 million. Due to delayed implementation, the total amount spent in 2002was � 7.84 million.

Refundable tax credit households in less favoured areas. This scheme is aimed at low-income householdsresident in mountainous and other less favoured areas. More specifically, the credit amounted to � 600 per yearfor households with annual family income below � 1,500 and to � 300 per year for households with annualfamily income of up to � 2,200. The relevant amount was credited against future tax liability. The scheme wasexpected to benefit 140,000 families at a cost of � 70.43 million, but only � 21.02 was spent in 2002.

Social insurance contributions rebate for low earners (since August 2000). Workers earning 100–105.5 percent of the minimum wage are entitled to a full refund of their IKA social contributions for pension (6.67 percent of wage). The monthly value of the refund was up to � 38.61 in 2002, when the scheme benefited 31,000workers at the cost of � 15.5 million.

Part-time in-work benefit (since April 2001). Those out of work for twelve months or more taking up a part-time job for at least four hours daily can apply for this benefit of � 88.04 a month for a period of up to twelvemonths. The benefit was conceived as an incentive for the reintegration of the long-term unemployed in thelabour market. At the time of its introduction it was expected that approximately 2,000 people would claim thebenefit at a total cost of � 4.25 million per year. No data on take-up are available.

administrative infrastructure, allowing for a gradual expansion of the social safety net towards “the ultimatecommon aim of a substantial provision of social guarantees” was not supported by hard data and wasactually disproved by what evidence existed, for instance, on the extremely low take-up of the measuresannounced in the previous National Action Plan of two years before. This issue, largely glossed over in the2003 National Action Plan, is discussed in more detail below.

On the topic of poverty and social benefits, the National Action Plan pointed to the substantial realimprovements in the value of the basic pension for farmers and the pension supplement EKA� (20.7 percent and 63.8 per cent respectively in 2000–03). As a result of that, provisional data showed that the risk ofpoverty for those over 65 was 32.9 per cent in 2001 and 30.1 per cent in 2002, compared to 31.2 per cent in2000 and 33 per cent in 1999 (the corresponding figure for the EU-15 as a whole was 17 per cent).27

Other than that, not a single new benefit was foreseen for the 2003–05 period. It looked as if the debacleof the measures introduced in 2000–01, eventually shown to have reached only a tiny fraction of theintended beneficiaries, had taken the much celebrated strategy of selective provision into a blind alley.

Overview of anti-poverty policies

Where does all that leave us in terms of anti-poverty policies?28 Table 2.10 shows selected benefits in termsof fiscal cost, coverage and adequacy. For the sake of comparability, large-scale social insuranceprogrammes such as minimum pensions, farmers’ basic pensions, invalidity pensions, etc. are left out of thisoverview, in spite of the fact that they have an obvious— though not clearly identifiable—“assistential”dimension. Contributory family allowances are also omitted. In contrast, the new refundable tax credits areincluded in the analysis, even though formally not part of social security.

On the whole, broadly defined social assistance accounts for a small but not entirely negligible part of thesocial protection system. Overall spending reached � 1,683 million or 1.2 per cent of GDP in 2002. In terms

44 MANOS MATSAGANIS

of coverage, the total number of direct recipients (in other words, not counting remaining members ofreceiving households) was almost 1.3 million or 11.7 per cent of the population. On the other hand, socialassistance scored low in terms of adequacy; even relatively generous benefits such as social pensionsamounted to � 156 monthly, a replacement rate of 28.5 per cent with respect to the statutory minimum wage.The corresponding figures for all benefits combined were � 140 and 25.5 per cent respectively.

From the point of view of comprehensiveness, the variety of schemes currently in force, including thoserecently introduced, constituted a frag

Table 2.10 Selected anti-poverty benefits in Greece (2002)

Number of recipients Expenditure (� million) Benefit rates (� per month)

Unemployment and in-workbenefits

290,714 442.15 239.82

Unemployment insurance 259,003 426.22 264.39

Unemployment assistance 711 0.43 150.00

Contributions rebate for lowearners

31,000 15.50 36.60

Old age benefits 405,000 489.14 102.40

Social pension to uninsuredelderly

40,000 108.22 156.13

Pensioner solidaritysupplement EKA�

365,000 380.92 96.51

Disability benefits 138,750 305.28 183.35

Severe physical disabilitybenefit

73,150 159.05 181.00

Other disability benefits 65,600 146.23 185.75

Family benefits 339,399 377.82 100.70

3rd child benefit 38,342 66.39 141.03

Large family benefit 82,008 106.80 149.51

Pension to mother of manychildren

183,584 185.78 81.09

Unprotected child benefit 34,440 18.19 44.02

Single parent benefit 225 0.31 115.19

Birth grant to uninsuredmother*

800 0.35 440.20

Housing benefits 37,578 35.41 125.06

OEK rent subsidy(contributory)

36,768 34.17 125.00

Housing benefit touninsured elderly

810 1.24 127.96

Emergency benefits* 11,006 4.40 207.72

Refundable tax credits 60,000 28.86 40.08

Tax credit fordisadvantaged areas

47,000 21.02 37.27

GREECE: ANTI-POVERTY POLICY 45

Number of recipients Expenditure (� million) Benefit rates (� per month)

Tax credit for schoolchildren

13,000 7.84 50.26

All benefits 1,282,447 1,683.06 139.95

% population 11.7 – –

% of GDP – 1.2 –

% min. wage – – 25.5

NotesIn the case of family benefits, number of recipients is number of claiming units (i.e. families not children). Unemployment

insurance is provided for an average of 7 months. Old age benefits are paid 14 times a year. Birth grants andemergency benefits, denoted by (*), are granted as a lump sum. Average monthly benefit rates are weightedby number of recipients, exclude lump sum benefits, but take no account of variations in the number ofmonthly payments per year. The figures on emergency benefits are tentative. The daily minimum wage (� 21.95) was assumed to be paid 25 times a month.

mented structure that left a large number of low-income families unprotected. A partial list of coverage gapsincluded most of the long-term unemployed whose eligibility to benefit has been exhausted, new entrants tothe labour market ineligible for any cash support, those precariously employed with no social insuranceentitlements to draw upon in the event of temporary loss of earnings, low-earning workers with less thanthree children, owners of unprofitable small businesses or farms, retired farmers on the basic pension alone,widows with no retirement benefits of their own right, and persons incapable of working with no access to adisability benefit or invalidity pension.

As analysed in the rest of this section, in the absence of a policy shift towards strengthening the socialsafety net and investing in administrative infrastructure, coverage gaps were likely to persist in spite of theprovisions of the two National Action Plans for Social Inclusion.

Unresolved issues

Having rejected the option of gradually moving towards a guaranteed minimum income scheme, recentgovernment policy has favoured a strategy of filling gaps in the social safety net. It attempted to do so byraising the value of some benefits (primarily EKA� and the social pension) over and above the rate ofinflation, and by introducing certain new measures targeted at particular groups.

On paper, the latter seemed to go some way towards strengthening the social safety net. In particular, thatwas the case of unemployment assistance for older workers, widely seen as an important step in the rightdirection. However, such hopes were quickly dashed. The May 2001 National Action Plan for SocialInclusion had announced that the scheme would be launched in January 2002 and anticipated that 35,000unemployed workers would claim, at a cost of � 59 million annually. Instead, quite incredibly, the Ministryof the Economy and Finance at first failed to make the necessary provision in the 2002 budget and thenagreed to earmark a mere � 15 million. As it turned out, one-tenth of that amount would have beensufficient: ten months after its introduction, the new benefit was being paid to exactly 711 individuals, or 2per cent of those originally expected to claim.

A variety of factors was to blame for such poor performance. From the start, design faults were evident.Specifically, the maximum duration of benefit, at twelve months, was too short by European standards for apolicy of this type. Moreover, there was little justification for imposing an age restriction (over 45), giventhat an income test applied as well. The implicit assumption that younger unemployed workers all lived in

46 MANOS MATSAGANIS

well-off families beggared belief. As for the income threshold for eligibility (� 4,174 a year for a family offour), it was clearly too low and was subsequently raised by a further � 2,000. On the other hand, as onvarious other occasions discussed later, the level of benefit was not reduced as income rose but withdrawnaltogether on crossing the threshold, while no provision was made for the partial disregard of earnings toencourage a return to work. Administrative difficulties such as the patent lack of co-ordination betweenfunding and spending ministries within the government or the insufficient preparation on the part of the“Manpower Employment Organisation” (OAE� ) helped make the launch of this promising policyinstrument a rather hapless event.

Equally disappointing was the earlier case of the social contributions rebate to low earners. Soon after thescheme was introduced in August 2000, reports began to come in that participation was lower than expected.The matter became public in February 2001, at which point an internal inquiry of the Ministry of Labourand Social Security revealed that of the 470,000 workers expected to participate, only 40,000 actually did so(i.e. a non-take-up rate of 91.5 per cent). The inquiry identified various design flaws and administrativeproblems. Benefit rules required that employers pay eligible employees a net wage plus the rebate and thenpay the social insurance agency IKA the usual contributions minus the rebate. Therefore, the scheme largelydepended on the ability and willingness of employers to carry out the necessary procedure and then pass onthe benefit to eligible employees. Yet, experience showed that many employers were unable or unwilling todo so. In view of that, it was decided that thereafter the rebate would be administered as a refundable taxcredit. Nevertheless, three full years after its launch, participation to the scheme had declined further to 31,000.29

The decision to entrust the social contributions rebate to the income tax system was part of a wider policyshift to redistribution via fiscal instruments. This shift is bound to raise serious issues. To start with,whatever the relative weight attached to fiscal instruments, these cannot substitute for a well-functioningsocial administration, currently lacking. Administrative weaknesses, often cited to explain the institutionaldivergence of welfare in Greece vis-à-vis the rest of the EU, will not be solved unless tackled. On the otherhand, tax credits by design fail to benefit those in greatest need. Most low-income families are legallyexempted from the obligation of filing an income tax declaration. Even those who do qualify for assistancewill receive the credit after income tax declarations are processed, that is 8–10 months from the end of thefinancial year in which low income was reported. Due to these defects, tax instruments are generallyunsuitable for the pursuit of anti-poverty objectives, as the experience of all other EU countries confirms.This is not to say that such instruments cannot be used for broader redistributive purposes further up theincome scale.30 Third, the tax credits announced in the 2001 National Action Plan are likely to compoundfragmentation. In keeping with tradition, but for no apparent reason, they were selectively aimed at householdsin less favoured areas and at families with children aged 6–16. As a result, households in more favouredareas and families with younger children remained ineligible for assistance, even if poor. Moreover, at � 2,200 and � 3,000, respectively, the annual income thresholds were so low that they excluded many poorfamilies within the narrow categories chosen by the government. On a final note, the tax credit for childrenaged 6–16 was also inconsistent with the already existing (non-refundable) tax credit for children,reinforcing the general incongruity of income support measures with each other.

It ought to be clear that the issues, briefly raised above, of poor take-up, flawed design, administrativeproblems, etc. are not restricted to the policies introduced in recent years. On the contrary, these are part ofa more general pattern that is worth discussing more systematically and in somewhat more detail.

GREECE: ANTI-POVERTY POLICY 47

Benefit delivery

In the absence of an automatic system of targeted transfers, the availability of an efficient administrativestructure for the provision of benefits to potential claimants becomes a necessity. On this count, theperformance of Greek benefit agencies leaves much to be desired.

In general, benefits are administered by agencies not originally established for the purpose. For instance,the “many-children” benefits are rather incongruously provided through O� A, the Agricultural InsuranceOrganisation. The requirement of timely provision of assistance is as a rule not satisfied: benefits are oftendelivered with great delay and arrears are common, while some schemes pay benefits once a year or everytwo months. Moreover, instead of an open system of applications to ensure that urgent needs can be metwhenever they arise, such periods are usually short and with strict deadlines; if missed, beneficiaries must waitfor another year.

In spite of the above, no effective procedure for addressing grievances was available until recently. Thecreation of the “Office of the Ombudsman” in 1998 has made it the main depository of complaints about thefailures of benefit agencies (30 per cent of all cases taken up in 2001). Limited information aboutentitlements was identified as one of the more crucial areas of concern.31

Poor information and haphazard administration erect a barrier between benefits and potential claimants.Policy-makers often lose interest in a new measure after it has been announced, while benefit-payingagencies do not normally take the view that it is their responsibility to ensure that cash assistance actuallyreaches those it is intended to help. As a result, take-up of benefits is not given serious attention, despiteearlier discussed evidence that it can be dispiritingly low.

In contrast, concerns about “leakages” to illegitimate beneficiaries are strong. As seen below, officialopposition to a minimum guaranteed income scheme is often explained along these lines. It is true that thereliability of methods to assess the incomes of potential claimants is highly questionable. Poor taxcompliance is known to constitute a serious obstacle.32 Moreover, a legal exemption from filing an incometax return operates below a certain threshold, affecting 1.4 million individuals or 13 per cent of the totalpopulation, as a result of which, tax data would be of little use in identifying the poor, even if they werecredible.

Nevertheless, such concerns equally apply to existing income-tested benefits: all agencies, with noexception, rely on signed statements supported by a copy of the previous year’s tax return. Indeed,reservations about the reliability of income tax records have not interfered with the introduction ofrefundable tax credits, nor have they prevented the decision to shift social policy tasks to the tax system.

Benefit design

Effective targeting also needs to conform to some minimum standards in terms of design. For instance, theamount of benefit should preferably be inversely related to the claimant’s income, either linearly or at leastin a “graduated” form, with higher rates corresponding to lower income bands. On the other hand, the rateof benefit reduction as income rises must be less than 100 per cent in order to avoid the “poverty trap”.“Income disregards”, or exceptions from income as assessed by the benefit agency, are also useful, playinga similar role in softening work disincentives. Moreover, income must be defined in a consistent fashion,while care must also be taken to assess incomes of existing recipients net of the benefit itself.

The above conditions may seem obvious, but the fact is that more often than not they are violated. AsTable 2.11 shows, most income-tested benefits are paid at a flat rate to all those who meet the eligibilitycriteria. That means that individual A with original income just below the threshold ends up better off than

48 MANOS MATSAGANIS

individual B with original income just above the threshold: A will receive the full rate of benefit while Bwill get nothing.

The pension supplement EKA� seemed likely to be an exception to the rule, as considerable effort hadbeen put into its design. EKA� featured four rates of benefit corresponding to an equal number of incomebands, carefully chosen to avoid inconsistencies. Nonetheless, successive revaluations of both benefit rateand income thresholds distorted the original design. As a result, pensioner B of Table 2.12 risked ending upworse off than pensioner A when both received EKA� , reversing their original position. Such an effectcontradicts common sense, not to mention optimal taxation theory, while constituting a disincentive to paypension contributions and an incentive to manipulate income tax returns.33

Moreover, the income concept varies between benefits and agencies, for no clear reason but withimplications for eligibility. Income allowances are rare and non-standardised. Some benefits are taxed asincome (a feature usually associated with universal benefits), while others are not. Finally, on threeoccasions the benefit itself is included in the definition of income used when assessing claims. This givesrise to the rather bizarre situation in which year 1 beneficiaries whose income from other sources remainsfixed risk losing the benefit in year 2, qualifying again in year 3, losing the benefit again in year 4, and soon.34

Table 2.11 Design features of selected benefits

Graduatedbenefita

Income conceptb Benefit included Incomedisregardsc

Benefit taxed

PensionersolidaritysupplementEKA�

yes taxable no none yes

Social pensionfor non-insuredelderly

no declared no benefits no

Unemploymentassistance

no presumptive no benefits yes

3rd child benefit no presumptive yes none yes

Large familybenefit

no presumptive yes none yes

Pension tomother of manychildren

no presumptive yes none yes

Unprotectedchild benefit

no total no rent no

OEK rent subsidy yes declared no savings yes

Notesa EKA� has a full rate and three reduced rates that correspond to four pension income bands (personal and family

incomes are also taken into account but there is a single income threshold for each). OEK rent subsidy isreduced euro for euro as income rises over a certain range. All other benefits are paid at a single rate andwithdrawn totally as soon as income crosses the threshold,

b Declared income is higher than taxable because various exemptions operate. “Presumptive taxation” applies to somegroups of tax payers (e.g. the self-employed) when their declared income is below a certain minimum,

c In the case of the social pension, the assessment excludes income from social assistance. In the case of unemploymentassistance, income from unemployment insurance benefit, maternity benefits and sickness benefits is

GREECE: ANTI-POVERTY POLICY 49

Graduatedbenefita

Income conceptb Benefit included Incomedisregardsc

Benefit taxed

disregarded. In the case of OEK, rent subsidy excludes income from interest on savings, family benefits,unemployment compensation (both ordinary benefit and severance pay), disability allowances, as well aspension payments in arrears. In the case of unprotected child benefit, the income concept includes privatetransfers and social benefits.

Benefit revaluation

Indexation of benefits, and of the relevant income thresholds, is a closely related issue. Here, the minimumrequirements are consistency and pre-dictability. On the contrary, as Table 2.13 shows, indexation is usuallyad hoc and sometimes skipped altogether. Revaluation policy is erratic, with some benefits receiving higherrises than others, for no apparent reason and with no explanation given. More often than benefit rates,income

Table 2.12 Inconsistencies of income tests: EKA�

Pension income pre-EKA� Value of EKA� Pension income post-EKA�

Pensioner A 5,262 1,146 6,408

Pensioner B 5,468 859 6,327

NoteFigures are annual incomes and benefit rates in euros in 2001.

Table 2.13 Indexation and revaluation of selected benefits (1996–2001)

Indexation Revaluation

No. of times skipped Real change in

Method applied Benefit amount Income threshold Benefit amount Income threshold

PensionersolidaritysupplementEKA�

inflation 1 1 108.1 9.0

Social pensionfor non-insuredelderly

ad hoc 0 0 63.8 63.8

Pension tomother of manychildren

inflation 1 3 −1.3 −12.5

Large familybenefit

inflation 0 3 14.5 −6.3

3rd child benefit inflation 0 3 16.2 −2.8

Unprotectedchild benefit

ad hoc 4 4 6.4 49.9

OEK rent subsidy ad hoc 2 1 11.3 0.1

IKA minimumpension

ad hoc 0 n.a. 11.6 n.a.

50 MANOS MATSAGANIS

Indexation Revaluation

No. of times skipped Real change in

Method applied Benefit amount Income threshold Benefit amount Income threshold

Unemploymentinsurance

min. wage 3 n.a. 3.0 n.a.

Disabilityallowances

ad hoc 0 n.a. 37.0 n.a.

NotesThe length of the period examined in the “no. of times skipped” column is five years (1996–2001) except for the

pension to mother of many children, large family benefit and 3rd child benefit (four years), as these becameincome tested in March 1997. The same goes for EKA� , which was introduced in mid-1996 and was firstrevalued in 1998. The income threshold in the case of the social pension (for a two-person household) is thefarmers’ basic pension (if received by the claimant’s partner). IKA minimum pensions, unemploymentinsurance benefit and disability allowances are not income tested. Disability allowances are proxied bysevere disability allowance, claimed by over 55 per cent of all disability allowance recipients.

thresholds fail to keep pace with inflation or are left unchanged in nominal terms, so that beneficiaries withfixed real income risk crossing the threshold and thus being disqualified.

“Generous selectivity”?

Targeted benefits differ in terms of how severe the income test is. There clearly is a great differencebetween stringent means tests aimed to restrict assistance to the poor and “affluence tests” intended merelyto exclude the rich.35 In fact, recent policy has been justified in terms of a superiority of “generousselectivity” over “miserly universality”.36 It may therefore be interesting to see exactly where along thatspectrum the Greek case lies. Table 2.14 gives the position of income thresholds for access to variousbenefits in relation to a survey-derived poverty line set at 60 per cent of median equivalent income.

While a clear pattern is difficult to discern, benefits fall into three groups. First, the “many-childrenbenefits” (3rd child benefit, large family benefit and pension to mother of many children), designed toexclude the highest incomes, until the income test was abolished in 2002. Second, EKA� and OEK rentsubsidy, granted on the basis of more restrictive income criteria, with income thresholds for most categoriesof claimants just over the poverty line of 60 per cent of median. Third, all remaining

Table 2.14 Income thresholds as percentage of the poverty line (2002)

Single Couple (no. of children)

0 1 2 3 4 5

PensionersolidaritysupplementEKA�

124 128 107 92 – – –

Socialpension fornon-insuredelderly

39 26 21 18 – – –

GREECE: ANTI-POVERTY POLICY 51

Single Couple (no. of children)

0 1 2 3 4 5

Unemploymentassistance

52 41 40 39 – – –

Unprotectedchild benefit

– – 28 26 24 23 22

OEK rentsubsidy

151 108 104 101 99 97 96

Tax credit toschoolchildren aged 6–16

– – 29 25 22 – –

Tax credit tofamilies inremote areas

39 26 22 18 16 – –

3rd childbenefit

– – – – 180 – –

Large familybenefit

– – 300 257 225 200 198

Pension tomother ofmanychildren

189 126 – – – – –

NotesThe poverty line was assumed to be 60 per cent of median equivalent income, adjusted for family size with the OECD

modified equivalence scale also used by Eurostat (which assigns a value of 1.0 to the household head, 0.5 toother adults and 0.3 to each child). The poverty line for a family of four in 2002 was � 992. The figures forEKA� were derived by taking the personal income threshold for claimants living alone and the familyincome threshold for all others. The income test for access to the so-called “many-children benefits” (3rdchild benefit, large family benefit and pension to mother of many children) was abolished in 2002. As aresult, the income thresholds reported above have been calculated with respect to the poverty line of 2001(� 951 for a family of four).

benefits, including the ones introduced with the National Action Plan 2001–03, targeting exclusively thepoorest members of the relevant categories.

The example of the social pension, the only benefit coming close to the definition of a last resort safetynet in Greece, is illustrative—and rather sobering, given the overall generosity of the rest of the pensionsystem. For the social pension to be awarded to one member of an elderly couple, their resources must beless than the equivalent of 26 per cent of the poverty line. Surely not the finest example of generous selectivity.

As a matter of fact, the absence of a last resort benefit (targeted in nature but universal in scope) remainsa crucial missing link in Greece’s social safety net. Given the focus of this book on minimum income schemes,the following section reviews the recent political debate on minimum income, presents a proposal for theintroduction of such scheme in Greece and discusses the requirements for this reform to be successful.

52 MANOS MATSAGANIS

Minimum income as a contested issue

The political debate on minimum income

The history of guaranteed minimum income as a policy issue for Greece is short. Just over a decade ago,Council Recommendation 92/441 on common criteria concerning sufficient resources and social assistancein social protection systems failed to make any impression on the domestic policy debate, as it passedlargely unobserved. As late as 1998, a periodic evaluation by the European Commission of progresstowards implementation of Council Recommendation 92/441 mentions Greece only once and only to saythat no minimum income scheme at all operates there, neither at a national nor at a local level.37

Things began to change towards the end of the 1990s. The experimentation with minimum income schemesin Italy and Portugal posed the question of whether Greece should follow suit. After all, the three countriesshared features such as extended families and informal employment that, as conventional wisdom had it, wereincompatible with the introduction of guaranteed minimum incomes. Furthermore, all three were at thattime ruled by centre-left majorities. As a result, the subject began to attract some interest in the policydebate within Greece. This was given further impetus by the decision at the conclusion of the Portuguesepresidency of the EU in the first half of 2000 to set in motion the “Lisbon process” that requires thatmember countries, starting from 2001, submit biennial National Action Plans for Social Inclusion.

A blueprint for introducing a guaranteed minimum income scheme in Greece was actually presented toPrime Minister C.Simitis in January 2000.38 The proposed scheme, in line with standard policy,combined financial support in case of acute hardship with individual social reintegration plans. At the time,the proposal was set aside after brief consideration, mainly on the grounds of concerns about“uncontrollable budgetary costs” at a time when meeting the Maastricht criteria for entry to EMU was, quiterightly, the Greek government’s overriding aim.

In the run-up to the general election of April 2000, the ruling socialist party included a pledge in itsmanifesto to create a “Network against Poverty and Social Exclusion”. After re-election, the Prime Ministerasked the new Minister of Labour and Social Insurance to convert the pledge into an operational plan. TheMinister then set up an informal committee to assess various options, including that of a guaranteedminimum income. The task of setting out the goals of the “Network” expanded to include preparation of thefirst National Action Plan for Social Inclusion. As explained earlier, the National Action Plan for 2001–03indirectly ruled out the option of a guaranteed minimum income in Greece, a position made explicit in thesecond National Action Plan for 2003–05.39

Official opposition to guaranteed minimum income varied in resolution. It seemed to harden when a formerclose collaborator of the Prime Minister tabled a draft bill to introduce a guaranteed minimum income on 6December 2000. The motion, in effect a translation of Portuguese legislation, was signed by a total of fifty-two socialist MPs (one-third of all), most of them loyal supporters of the government acting in good faith. ThePrime Minister reacted angrily to what clearly was an attempt to pre-empt government policy, but byattacking the proposal itself he silenced all internal discussion on an issue that had until that point been openfor decision.

Nevertheless, a month later, on 11 January 2001, all major newspapers carried almost identical reportssuggesting that the day before a high-level meeting chaired by the Minister for National Economy haddecided to regain the initiative on the issue. The reports, clearly the result of off-the-record briefing,indicated the Minister’s intention to work with the Minister of Labour and Social Insurance and theMinister of Health and Welfare on draft legislation with the aim of introducing a guaranteed minimumincome from 1 January 2002.

GREECE: ANTI-POVERTY POLICY 53

Almost immediately, on 13 January 2001, the government seemed to backtrack: while reiterating theircommitment to that aim in principle, “sources close to the Minister” began to express doubts as to whetherit was feasible in practice. At the same time, the leader of the opposition centre-right New Democracyannounced his intention to submit a draft law on a minimum income scheme.

On 19 February 2001, at a conference organised by New Democracy, the Minister of Labour and SocialSecurity painted in dark colours the budgetary implications of a move towards a minimum income scheme,while the leader of New Democracy, C.Caramanlis, explained his party was evaluating the possibility ofendorsing the introduction of a guaran teed minimum income as official party policy. The earlier promiseddraft law was, after all, not to be submitted just yet.

As the deadline for the first National Action Plan for Social Inclusion drew closer, in May 2001, it becameclear that the government had no intention of preparing the ground for a minimum income reform. At thispoint, interest in the issue seemed to recede once again. Indeed, as Parliament deliberated on the contents ofthe Republic’s new Constitution in 2002, a proposal by the Coalition of the Left Party to make access to aminimum income a constitutional right was rejected by both the socialist majority and the conservativeopposition.

The issue made headlines again in August 2003, as the government prepared its autumn offensive in alast attempt to reduce the lead enjoyed by New Democracy, according to opinion polls, a mere eight monthsaway from the general election scheduled to take place in April 2004.40 Newspaper reports of 24 August2003 suggested the Ministry of the Economy and Finance was re-assessing the proposal presented to thePrime Minister back in January 2000, as a possible centre-piece of the much-hyped “social package” to beannounced in September. As explained earlier, the government opted instead for a further significantincrease of the flat-rate farmers’ pension and the pension supplement EKA� .

Meanwhile, the opposition New Democracy was reported to be hesitating between the temptation ofmatching the government’s “social package” with a firm pledge to introduce a guaranteed minimum incomewhen in government, and the perceived need to adopt a prudent fiscal stance. The party’s emeritus chairmanand former Prime Minister C.Mitsotakis summed it up well when on an interview published on 21September 2003 he described the statutory right to a minimum income as a noble ultimate aim, while rulingit out as a practical possibility in the foreseeable future.

On the whole, the idea of introducing a guaranteed minimum income in Greece seems to divide politicalelites. A recent journalistic inquiry on political attitudes to minimum incomes published on 26 October 2003confirmed that the issue cut across political party boundaries. A.Diamantopoulou (the EuropeanCommissioner for Employment and Social Affairs and a former socialist minister) came out cautiously infavour, as did the conservative shadow minister and the social policy spokesman of the Left Coalition. Onthe contrary, the Minister of Labour and Social Insurance as well as the chairman of the trade unionsconfederation � � EE, both socialist, were strongly against the proposal to introduce a guaranteed minimumincome.

But was it really the case that introducing a guaranteed minimum income scheme would haveundermined Greece’s efforts to meet the Maastricht criteria in the late 1990s and would rock the boat in thenot-so-tranquil waters of the Stability Pact in the rest of this decade? The next paragraph briefly reviews thefindings of the only study so far actually attempting to quantify the fiscal and other implications of thiscontested policy proposal.

54 MANOS MATSAGANIS

Simulating minimum income

What would be the effects of introducing a guaranteed minimum income scheme in Greece? While it is truethat to know the full answer we will have to wait until one is really introduced, the possibility of a satisfactoryapproximation before that time does exist. The best available technique is known as “microsimulation” andconsists in examining the income distribution of a representative sample of the population as it actually iswith what it would be if the policy change in consideration were in force.

In such a simulation exercise, reported elsewhere,41 a household would be eligible for benefit if itsreference income were less than the minimum income guarantee that applied to that household. The benefitpaid to eligible households was just enough to bring their reference income up to the level of the minimumincome guarantee.

To account for the inevitable presence of targeting errors, the scheme’s effects were simulated under varyingassumptions. In general, targeted programmes are subject to errors of two types.42 Type I errors (“falsenegatives” or non-take-up) imply the non-provision of benefit to eligible applicants; type II errors (“falsepositives” or leakages) refer to the provision of benefit to non-eligible applicants. Therefore, while perfecttargeting (i.e. full take-up, no leakages) was the baseline scenario, alternative estimates involvingcombinations of leakages and non-take-up were also presented.

Table 2.15 suggests that under perfect targeting approximately 700,000 persons (6.4 per cent of Greece’spopulation) would be expected to participate in the minimum income scheme. The cost of transfers underthe scheme would be � 269 million (0.23 per cent of the country’s GDP in 2000). Non-take-up and leakageshad opposing effects, to a certain extent, cancelling each other out. Moderate targeting errors (20 per centnon-take-up and 50 per cent leakage) had similar outcomes to the baseline scenario in terms of bothparticipation and cost, though clearly inferior in terms of anti-poverty effectiveness.

As a matter of fact, the greatest reduction in poverty was shown to be achievable under conditions ofperfect targeting. Even though leakage to households over the threshold for eligibility to minimum incomehad a positive distributional effect if these households were below the conventional poverty line of 60 percent of median, non-take-up had a far stronger negative effect, as it affected households further down theincome scale (the “extremely poor”). In view of that, the presence of both non-take-up and leakages, whileneutral in terms of recipient numbers and fiscal cost, weakened the anti-poverty impact of the minimumincome scheme in comparison to the baseline scenario of perfect targeting.

Table 2.15 Effects of minimum income under alternative scenarios (2000)

Target efficiency Participation Cost

non-take up(%) leakage(%) number % population � million % GDP

10 25 678,300 6.2 254 0.21

10 50 787,700 7.2 297 0.25

10 100 918,900 8.4 329 0.28

20 25 612,600 5.6 231 0.19

20 50 722,000 6.6 276 0.23

20 100 842,300 7.7 308 0.26

30 25 557,900 5.1 208 0.18

30 50 667,300 6.1 253 0.21

30 100 798,600 7.3 285 0.24

Perfect targeting 700,100 6.4 269 0.23

GREECE: ANTI-POVERTY POLICY 55

Target efficiency Participation Cost

non-take up(%) leakage(%) number % population � million % GDP

NotesEstimates derive from Matsaganis et al. (2001), corrected for population and GDP aggregates made available later.

Non-take-up is defined as the number of eligible persons not participating to the scheme as a proportion ofall eligible persons in the perfect targeting scenario. Leakage is defined as the number of non-eligiblepotential beneficiaries as a proportion of conceal a proportion of their income with a view to passing themeans test. That proportion was set equal to 100 per cent of all non-work non-transfer income and 50 per centof income all eligible persons in the perfect targeting scenario. Potential beneficiaries were assumed to fromwork if earned in the private sector, while income earned in public employment or from transfers wasassumed to be declared in full. If the resulting transfer was positive as a result, the persons concerned werethen included in the leakage group.

Future prospects for minimum income in Greece

The decision to introduce a minimum guaranteed income programme in Greece is likely to hinge upon twoclosely linked considerations: budgetary implications and administrative requirements.

As shown above, the total cost of transfers provided under the minimum income programme wasestimated at 0.23 per cent of GDP (approximately 1 per cent of all expenditure on social protection). Thisfinding is in line with the experience of other EU countries with minimum guaranteed income programmesin operation. Moreover, the estimate remains robust under quite severe assumptions of targeting errors.

Nonetheless, policy-makers are unlikely to be reassured by cost estimates derived from family budgetsurveys, unless they find confirmation in the administrative databases that would be used if the programmewere to be actually implemented. As discussed earlier, tax records remain unsuitable for the purpose ofassessing claimants’ income. On the one hand, tax evasion has been somewhat contained recently but by nomeans defeated. On the other hand, households with very low income are exempt by law from theobligation to submit income declaration forms for taxation purposes.

Moreover, as shown before, social administration in Greece lacks a tradition of providing income-testedsupport. The recent attempt to provide each citizen with a single social security number, storing allinformation on benefits claimed in computerised individual records, has run into serious difficulties.Administrative problems raise the risk that the authorities responsible for a guaranteed minimum incomescheme might be inundated with false claims. To minimise such a risk, these and other problems wouldhave to be sorted out before the scheme can be successfully implemented.

Nevertheless, it is important to remember that minimum income programmes are not pure cash transfers.Standard rules require recipients to be available for work or training—although exceptions may apply incase of illness, disability or caring responsibilities for children, the disabled, frail elderly, etc. Combiningincome support with “reactivation” measures is likely to prove more demanding from an administrativepoint of view. However, such requirements have a valuable role as a screening device, deterring thosetempted to apply for assistance while holding a job in the informal sector.43

Obviously, the main advantage of properly designed reintegration programmes is the promise to enablethe recipients of state assistance to escape poverty rather than stay dependent on benefits. The “exit” ofbeneficiaries from social assistance into paid employment has the added effect of reducing expenditure onminimum income as well as on other benefits.44

From this point of view, the development with ESF support of vocational training programmes and, morerecently, “accompanying support measures” represents a resource that could be drawn upon in the event of a

56 MANOS MATSAGANIS

minimum income scheme being implemented in the future. For instance, eligibility criteria for participationin the various reactivation programmes already in operation could easily be modified to allow priorityaccess for minimum income recipients.45

On a rather different note, the decision to set up a guaranteed minimum income programme is likely torest on political considerations as well. In general, the essentially mild interest that politicians, the mediaand public opinion mostly reserve for issues of poverty or social exclusion in Greece can be explained byreference to the political economy of welfare. Poverty and social exclusion directly involve no more than asmall minority of population. The socially excluded are often invisible and sometimes politically excludedas well, as in the case of foreign immigrants. Therefore, their voice carries little weight with political parties,trade unions and the press. As a result, the plight of, for example, immigrant children dropping out ofschool, or of foreign workers living in overcrowded basements, is met at best with helpless resignation. Incontrast, speculation that the government’s pension reform plans may make it slightly more difficult for publicsector workers to retire in their early fifties can be relied upon to provoke waves of indignation throughoutthe country.

In a sense, the heart of the matter seems to be a failure of representation. Both organised interests(including trade unions) and political parties (including those of the left) seem more sensitive to thegrievances of the liberal professions and of the public sector workers who make up the bulk of theirmembership, than to the aspirations of precarious workers, foreign immigrants, inactive women, the youthand others who are not involved in public affairs or do not even have the vote. As a result, the politicalagenda is shaped far more by the former than by the latter.

Pressure from genuinely concerned individuals or groups from the ranks of the comfortable majority, aclassic remedy to the emerging “culture of contentment”, is not entirely absent, however. A generalpresumption in favour of social justice plays a clear role in internal discussions within the unions and theruling socialist party. The same goes for organisations of the left, marred by a tendency to use poverty toscore points off the government. Independent commentators, journalists and academics have also played apositive, though limited, role. The rising profile of voluntary organisations, particularly active in theprovision of social care, is a promising development that could strengthen the case for a minimum incomein the future. The slow emergence of a civil society is partly balanced by the retreat of the Greek OrthodoxChurch away from Christian charity into the bitter politics of ethnic identity. Last, but far from least, theinfluence of the EU and the gentle pressure exerted under the open coordination method and the NationalAction Plans for Social Inclusion force opponents of minimum income to argue their case more openly thanthey would have wished.

What is still missing is a reform programme: a set of ambitious but realistic ideas, capable of forgingthese disparate forces into an “advocacy coalition”. Given that “the raw material of every reform is alwaysan ideal blueprint for change”,46 this is a serious deficit. Although this is not the ideal place for a detailedexposition of the variant of such a reform programme favoured by this author, the temptation to concludethis chapter with a few remarks on the main priorities is difficult to resist.

Conclusion

As discussed in the preceding pages, the “danger that some groups experiencing poverty may not be eligiblefor income support”47 remains largely undiminished, despite recent improvements. Clearly, a moresustained effort towards a coherent policy response to poverty and social exclusion is required. Thefollowing notes may be read as a contribution to the search for policy solutions.

GREECE: ANTI-POVERTY POLICY 57

Tightening the social safety net remains a basic priority. Policies to that effect include the extension ofunemployment assistance to all long-term unemployed and first-time job seekers with low income, theconsolidation of the various family allowances and many-children benefits into a flat-rate child benefitpayable from the first child, the unification of existing disability benefits into a single benefit withsupplements for extra needs due to disability, the strengthening of social housing policies and thedevelopment of housing benefits on a non-contributory basis, as well as the gradual incorporation ofminimum pensions, social pensions and EKA� into a uniform system of income support to the low-incomeelderly. The formal separation of social assistance from social insurance, with the former being financed outof general taxation and the latter from contributions, will provide a more favourable framework in whichalternative options can be assessed.

Cash benefits alone cannot work without a comprehensive range of services —especially in a transfer-heavy social protection system such as Greece’s. Social care for all those incapable of working (the elderly,the disabled) ought to be combined with active labour market measures to improve everyone else’s“employability”. Significantly, the promotion of a comprehensive strategy in which social security andactivation measures are regarded as complementary rather than alternative, requires a closer integration ofservices and transfers than is currently the case.

The general upgrading of administrative capacities is urgently needed. As seen earlier, the option of aguaranteed minimum income was rejected mostly on its alleged excessive administrative demands.Nonetheless, preciously little was done to address the chronically haphazard and ad hoc design and deliveryof social benefits. Most crucially, low take-up is a long-neglected issue that must be squarely faced. No safetynet, however well designed, can be truly effective unless a serious effort is made to ensure that benefitsreach those they intend to help.

The reformulation of social protection beyond the insurance principle with a view to strengthening socialcitizenship is a further requirement. A guarantee of access to a set of basic social rights—e.g. in health,housing or child care—regardless of contributory record will provide a floor on which targetedinterventions can then be built. The extension of basic rights to all foreign immigrants who pass a simpletest (e.g. proof of residence and no criminal record48) is key to an effective policy against poverty and socialexclusion.

The implementation of a guaranteed minimum income scheme will, as shown earlier, provide a focus fora concerted anti-poverty policy effort. By design, such a scheme will combine financial support in case ofextreme hardship with individual social integration plans, aimed at providing recipients with skills that enablethem to escape both poverty and dependency on benefits. The administrative difficulties involved inintroducing such a scheme in a country like Greece must not be under-estimated. Nevertheless, theexperience of other countries shows that seriousness of purpose and dedication to the task can remove the mostresistant of obstacles—at a modest cost.

Notes

1 Gough (1996:13). See also Ferrera (1996).2 For a recent review of minimum incomes in southern Europe, see Matsaganis et al. (2003).3 The classic text offering a concise review of modern Greek history is Svoronos (1972). The standard reference on

social policy in Greece between the two world wars is Liakos (1993), with the relevant sections usefullysummarised in English in Liakos (1996). For post-war developments, see Sotiropoulos (2003), Petmesidou(1996), Venieris (1996), Katrougalos (1996). On developments since 1980, see Guillén and Matsaganis (2000).

58 MANOS MATSAGANIS

4 In 1922 Greece had been continuously involved in military conflicts for ten years: the two Balkan Wars of 1912–13 were succeeded by the First World War, followed by the disastrous, as it turned out, Asia Minor Campaign.The Versailles Treaty had given Greece control over Smyrne (Izmir) and surrounding Aegean coast territoriespredominantly populated by Greeks. Eventually, the Greek army used these territories to launch an assault againstthe newly created Turkish Republic and its capital Ankara. In the ensuing backlash, the victorious Turks burnedSmyrne and “expelled between 1.3 and 1.5 millions of Greeks from Asia Minor, where they had lived since thedays of Homer.” (Hobsbawm 1992:133).

5 According to Liakos:

The ILO managed to promote the idea of social insurance in Greece through a fervent activitycomprising two visits by its director Albert Thomas and several interventions of his colleagues, andthrough an alliance with reformist trade unionists, politicians and high-ranking state officials.

(1996:96)

Pre-existing structures of what would now be termed social assistance were quite clearly inadequate.

According to an 1833 law, in case of indigence and certified incapacity to earn a livelihood, everymember of a commune was entitied to necessary assistance and support from the municipal budget.However, due to lack of resources, local government was never in a position to meet these needs.

(ibid.: 97)

6 The origins of “clientelist fragmentation” were in the pre-war period.

Between 1925 and 1940 many influential groups of self-employed workers succeeded in lobbying for apension fund of their own; for instance, lawyers, notaries, doctors, pharmacists, stockbrokers etc. In thesame period social insurance was also established for some categories of employees who either formed aneffective lobby or were exposed to comparatively greater risks.

(Liakos 1996:99)

7 According to Hobsbawm:

The only military regime actually backed by the USA in Europe was that installed in 1967 (probably onlocal initiative) by a particularly witless group of ultra-Right-wing Greek colonels in a country where civilwar between communists and their opponents (1944–49) had left bitter memories on both sides.

(1994:349)

8 On the tortuous path to pension reform in the 1990s, see Matsaganis (2002). For an analysis of a key episode in1996, see Featherstone et al. (2001).

9 For a more extended analysis, see Matsaganis and Tsakloglou (2001:191–196).10 For instance, a study of family contacts in the city of Piraeus in the late 1980s found that 56 per cent of young

couples had their children looked after on a regular basis by their parents or parents-in-law. Moreover, 36 per centreceived help with everyday household tasks, while as many as 35 per cent were given help in cash. See Maratou-Alipranti (1999:68).

11 For a recent review, see the collective volume of Southern European Society and Politics (winter 1998) on“Immigrants and the informal economy in southern Europe” (edited by Baldwin-Edwards and Arango). Theparticular experience of Greece is analysed in Fakiolas (1999), Sarris and Zografakis (1999), Baldwin-Edwardsand Fakiolas (1998).

GREECE: ANTI-POVERTY POLICY 59

12 Nevertheless, as reported at a conference on immigration organised in January 2003 by � � EE, the Greek tradeunions confederation, hostile attitudes towards immigrants on the part of administrators occasionally reaches thepoint of denying unemployment benefits to immigrant workers meeting all the conditions set down by law(Kapsalis 2003:4).

13 Social expenditure statistics were fully revised in 1998. Former efforts on the part of the National StatisticalService of Greece had omitted large items of expenditure and therefore seriously under-estimated total spendingon social protection. In response to this, the revision of older estimates and the annual production of new figureswere handed over to a committee chaired by a member of the Prime Minister’s Office. As a result of theexcessive (compared to elsewhere in Europe) politicisation of the issue, latest estimates may overstate the degreeof convergence of social expenditure in Greece relative to the European average. However, these revised figureshave been accepted and thereafter used by EUROSTAT.

14 Data for the year 2000. See Statistical Appendix I to NAPincl Greece (2003:3).15 For more detail, see Matsaganis et al. (2001).16 � EIIO� -MRC sample survey of 3,569 households in Greater Athens, carried out in Spring 1999.17 See Arapoglou (2002).18 For a fuller analysis of social benefits in Greece, see Matsaganis (2003b).19 For a detailed discussion of the origin and legal framework of disability benefits, see Amitsis (1994).20 The “mobility allowance” replaced the in-kind provision of free petrol for the converted vehicles of severely

disabled persons.21 A smaller-scale scheme is operated on a discretionary basis by the National Social Care Organisation (EOK� ). A

total of 250 children in 225 single-parent families were supported in 2001, at an annual cost of � 311,000.22 For a more complete description of cash assistance to families with children see Matsaganis (2001). For an

analysis of child poverty and assistance in southern Europe, see Matsaganis et al. (2004).23 See Heady et al. (2001). The study confirmed that the proportional decline in poverty due to cash transfers was in

the mid-1990s lower in Greece than in every other EU country except Portugal.24 For a systematic analysis, see Ferrera et al. (2002).25 NAPincl Greece (2001:7/19.)26 NAPincl Greece (2003:30–31).27 Appendix II: Laeken indicators to NAPincl Greece (2003:7).28 This overview refers to anti-poverty benefits, rather than to social assistance as such. According to the standard

definition, social assistance is aimed at individuals or families falling below a certain level of resources. See forexample Barr (1993:7–8). Eligibility is usually determined on the basis of a “means test” (a test of income plusassets). Other conditions of a categorical nature may apply as well, though a contributory record is normally notrequired. In well-designed schemes, the rate of benefit is negatively related to pre-transfer income (i.e. higherbenefit is paid to lower income recipients). Social assistance is typically funded out of general taxation. In thepolicy context of Greece, a whole series of departures from the standard definition of social assistance can beobserved. For instance, benefits are usually flat rather than inversely related to income, sometimes fundedthrough social contributions rather than being purely non-contributory, income tests are often implicit rather thanexplicit, and so on. Such departures render a meaningful use of the term almost impossible. In the light of recentchanges, relying on the criterion of the (at least implicit) presence of an income test, as proposed in Matsaganis(2000), would result in a significantly narrower list of benefits.

29 NAPincl Greece (2003:30).30 It is important to note that neither Working Tax Credit in Britain nor Earned Income Tax Credit in the USA

operate as last resort benefits. Both operate alongside conventional social assistance: Income Support in Britainand, more precariously, Temporary Assistance for Needy Families in the USA. Note also that advance paymentsof EITC are possible, on the initiation of the claimant. For a review of the new tax credits in Britain, see Brewer(2003), for an assessment of means-tested programmes in the USA, see the various contributions in Moffitt(2002).

31 According to the Ombudsman:

60 MANOS MATSAGANIS

The Office of the Ombudsman receives on a daily basis citizens’ complaints concerning the provision ofpoor or no information on th]e part of the administration. In this sense, inadequate information emerges asone of the greatest problems of administrative practice. On many occasions, citizens are constrained tolook for information from unofficial sources, with the result of being referred from desk to desk, unable toobtain authoritative information on the issue that concerns them. Yet, in an area such as social security,characterised by the technical nature and complexity of the relevant legislation, authoritative informationis a decisive factor for access to benefits, while conversely lack of information translates to de factonegation of citizen rights.

(2002:108)

32 A recent OECD report listed a variety of causes for this: the size of the informal economy (estimated at between24 per cent and 40 per cent of GDP), the large share of self-employment in the workforce, an inefficient taxadministration, the lack of a land register, the complexity and “continuous revisions and amendments” of taxlaws, the loopholes due to numerous tax allowances and exemptions, as well as “the extra burden of so-calledthird-party taxes that are extensively used to fund various institutions, e.g. the pension funds of lawyers,engineers and media workers”. See OECD (2001:93).

33 Incentives are complicated further by the fact that the other incomes taken into consideration (total personalincome and total family income) are not “graduated” but only have a single cut-off point.

34 The possibility of this “seesaw pattern” was drawn to my attention by Mr Spyros Yannopoulos (personalcorrespondence, 6 May 1998), Director of the Agricultural Insurance Organisation O� A.

35 Ferrera (1998b:87–93).36 For an analysis of the rise and fall of selectivity in Greece, see Matsaganis (2003a). 37 See CEC (1998:5).38 A shorter version was submitted to the committee responsible for the preparation of the National Action Plan for

Social Inclusion in March 2001. The two papers, still unpublished, were titled “Fighting poverty with aguaranteed minimum income: analysis of the proposal and outline of a plan for implementation” (30 January2000) and “Guaranteeing sufficient resources: alternative strategies for strengthening the social safety net” (14March 2001) respectively (both in Greek). The papers were drafted by this author, then an adviser to the PrimeMinister. Research on guaranteed minimum income included a visit to Portugal on the invitation of the Ministérioda Solidariedade e Segurança Social in October 1999.

39 See NAPincl Greece (2001:7) and NAPincl Greece (2003:30–32) respectively.40 Eventually the general election took place on 7 March 2004 and was comfortably won by New Democracy.41 See Matsaganis et al. (2001). The effects of a minimum income scheme were simulated for the year 2000 on the

basis of the following assumptions: (1) the annual amount of the minimum income guarantee for a singlerecipient was set at about � 1,800 a year (the level of the social pension in 2000); (2) the equivalence scale appliedwas that of the Portuguese Rendimento Mínimo Garantido, attaching a weight of 1.0 for each of the first twoadults, 0.7 for all additional adults and 0.5 for each child under 18; (3) farming households reporting annualincome below the level of the lowest insurance class under the new contributory old-age pension scheme(approximately � 3,600 in 2000) were assigned a reference income equal to that amount; (4) households headedby students were excluded, on the grounds that low-income students have access to in-kind benefits (free orsubsidised accommodation and meals) that provide a “functional equivalent” to a minimum income; and (5) anincome disregard of 20 per cent of earnings from work was simulated, as a means to soften the poverty trap,while incomes from other sources, including social transfers, were taken into consideration in full. ECHP data(1995 wave, uprated to 2000) were used for all estimations.

42 For a succinct discussion of imperfect targeting, see Atkinson (1998). For a survey of non-take-up of socialsecurity benefits in Europe, see van Oorschot (1991).

43 Activation is not necessarily identical to workfare and the Poor Law. For instance, the French revenu minimumd’insertion includes a reintegration clause that stops short of a work availability condition. For a benevolent viewof participation requirements as self-selection mechanisms, see Sen (1995: 18–19).

GREECE: ANTI-POVERTY POLICY 61

44 In the Netherlands, where activation policies were pursued with more enthusiasm, between 1994 and 1999employment increased by 900,000 people, even though the total numbers of hours worked rose by only 10 percent. See Hemerijck and Visse (2000). For a more sceptical view, see van Oorschot (2002).

45 On current rules, participants are entitled to a per diem subsidy if attending a training programme, but not if theyare referred to one of the accompanying measures. Given that the most disadvantaged categories are naturalcandidates for accompanying measures and that vocational training centres are often unwilling to sign them on(on the grounds that they are “difficult cases”), they are left with no access to either income support or training.Making income support conditional on low income rather than participation to a particular type of reactivationmeasure would be a sensible policy change quite irrespectively of whether it is seen in the context of minimumincome or not.

46 Ferrera (1998a:108). 47 CEC (2001:34).48 The requirement of legal residence for the eligibility of immigrants to benefits is certainly advisable—but it

crucially requires an effective legalisation procedure that is not always in place. In any case, emergency benefitscould arguably be extended to illegal immigrants, as in some EU countries.

References

Abramovici, G. (2002) Social Protection in Europe, Statistics in Focus Theme 3— 1/2002 Population and LivingConditions, Luxembourg: EUROSTAT.

Amitsis, G. (1994) “The Greek Social Assistance System: Main Dimensions and the Treatment of Specific TargetGroups”, National Report to the European Commission, Network on Minimum Income, unpublished manuscript.

Arapoglou, V. (2002) “The rhetorical management of poverty and homelessness in the National Plan for SocialInclusion”, Position paper 8 April, Arsis: Greek section of the European Federation of National OrganisationsWorking with the Homeless FEANTSA (in Greek).

Atkinson, A. (1998) Poverty in Europe, Oxford: Blackwell.Baldwin-Edwards, M. and Fakiolas, R. (1998) “Greece: the contours of a fragmented policy response”, South European

Society and Politics, 3 (3):186–204.Barr, N. (1993) The Economics of the Welfare State, 2nd edn, London: Weidenfeld & Nicholson.Brewer, M. (2003) The New Tax Credits, IFS Briefing Note 35, London: Institute of Fiscal Studies.CEC (1998) Report on the Implementation of the Recommendation 92/441/EEC of 24 June 1992 on Common Criteria

Concerning Sujficient Resources and Social Assistance in Social Protection Systems, Report of the Commission tothe Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions,COM (98) 774 final.

CEC (2001) Draft Joint Report on Social Inclusion, Communication from the Commission, COM (2001) 565 final (http://www.europa.eu.int).

Dennis, I. and Guio, A.-C. (2003) Poverty and Social Exclusion in the EU after Laeken (Part I), Statistics in FocusTheme 3–8/2003 Population and Living Conditions, Luxembourg: EUROSTAT.

Fakiolas, R. (1999) “Socioeconomic effects of immigration in Greece”, Journal of European Social Policy, 9 (3):211–230.

Featherstone, K., Kazamias, G. and Papadimitriou, D. (2001) “The limits of external empowerment: EMU, technocracyand reform of the Greek pension system”, Political Studies, 49:462–480.

Ferrera, M. (1996) “The ‘southern model’ of welfare in social Europe”, Journal of European Social Policy, 6 (1):17–37.

Ferrera, M. (1998a) Le trappole del welfare: uno stato sociale sostenibile per l’Europa del XXI secolo, Bologne: IIMulino.

Ferrera, M. (1998b) “The four ‘social Europes’: between universalism and selectivity”, in M.Rhodes and Y.Mény (eds),A New Social Contract?: Charting the Future of European Welfare, Basingstoke: Macmillan Press.

62 MANOS MATSAGANIS

Ferrera, M., Matsaganis, M. and Sacchi, S. (2002) “Open co-ordination against poverty: the new EU ‘inclusionprocess’”, Journal of European Social Policy, 12 (3): 227–239.

Gough, I. (1996) “Social assistance in southern Europe”, Southern European Society and Politics, 1 (1):1–23.Guillén, A.M. and Matsaganis, M. (2000) “Investigating the ‘social dumping’ hypothesis: welfare policies in Spain and

Greece in the 1980s and 1990s” , Journal of European Social Policy, 10 (2):120–145.Heady, C., Mitrakos, T. and Tsakloglou, P. (2001) “The distributional impact of social transfers in the European Union:

evidence from the ECHP”, Fiscal Studies, 22 (4):547–565.Hemerijck, A. and Visse, J. (2000) “Change and immobility: three decades of policy adjustment in the Netherlands and

Belgium”, West European Politics, 23 (2): 229–256.Hobsbawm, EJ. (1992) Nations and Nationalism since 1870: Programme, Myth, Reality, 2nd edn, Cambridge:

Cambridge University Press.Hobsbawm, EJ. (1994) Age of Extremes: The Short Twentieth Century 1914–1991, London: Michael joseph.Kapsalis, A. (2003) Conclusions of the General Confederation of Greek Workers Conference on a European

Immigration Policy, E� � µ � � � � � 92 2–9, Athens: Institute of Labour (in Greek).Katrougalos, G.S. (1996) “The South European welfare model: the Greek welfare state in search of an identity”,

Journal of European of European Social Policy, 6 (1):39–60.Liakos, A. (1993) Labour and Politics in Inter-War Greece: The International Labour Office and the Emergence of

Social Institutions, Athens: Research and Education Foundation of the Commercial Bank of Greece (in Greek).Liakos, A. (1996) “Welfare policy in Greece (1909–1940): from private needs to the social question”, Comparing Social

Welfare Systems in Southern Europe, vol. 3, 93–108, Paris: MIRE.Maratou-Alipranti, L. (1999) “Inter-generational relations in the modern era: theo ries, trends and practices”, Greek

Review of Social Research, 98–99:49–76 (in Greek).Matsaganis, M. (2000) “Social assistance in southern Europe: the case of Greece revisited” Journal of European Social

Policy, 10 (1):69–81.Matsaganis, M. (2001) “Income transfers to families with children”, Revue du Droit de l’Assurance Sociale, 6 (510):

414–422 (in Greek).Matsaganis, M. (2002) “Yet another piece of pension reform in Greece”, South European Society and Politics, 7 (3):

109–122.Matsaganis, M. (2003a) “The rise and fall of selectivity selectivity a la grecque”, in G. Standing (ed.), Minimum Income

Schemes in Europe, Geneva: International Labour Office, pp. 251–272.Matsaganis, M. (2003b) Muddling Through: Trials and Tribulations of Social Security in Greece, EUI Working Paper

SPS 2003/10, Florence: European University Institute.Matsaganis, M., Ferrera, M. Capucha, L. and Moreno, L. (2003) “Mending nets in the South: anti-poverty policies in

Greece, Italy, Portugal and Spain”, Social Policy and Administration, 37 (6):639–655.Matsaganis, M., O’Donoghue, C., Levy, H., Coromaldi, M., Mercader-Prats, M., Rodrigues, C.F., Toso, S. and

Tsakloglou, P. (2004) Child Poverty and Family Transfers in Southern Europe, working paper EM 2/04,Microsimulation Unit, University of Cambridge.

Matsaganis, M., Papadopoulos, F. and Tsakloglou, P. (2001) “Eliminating extreme poverty in Greece”, Journal ofIncome Distribution, 10 (1–2):40–57.

Matsaganis, M. and Tsakloglou, P. (2001) “Social exclusion and social policy in Greece”, in D.Mayes, J.Berghman andR.Salais (eds), Social Exclusion and European Policy, Cheltenham: Edward Elgar.

Moffitt, R. (ed.) (2002) Means-Tested Transfer Programs in the US, Chicago: University of Chicago Press.NAPincl Greece (2001) National Action Plan for Social Inclusion 2001–2003, Athens (http://www.europa.eu.int).NAPincl Greece (2003) National Action Plan for Social Inclusion 2003–2005, Athens (http://www.europa.eu.int).OECD (2001) Economic Surveys: Greece, Paris: Organisation for Economic Co-operation and Development.Ombudsman (2002) Annual Report 2001, Athens: Office of the Ombudsman (in Greek).Petmesidou, M. (1996) “Social protection in Greece: a brief glimpse of a welfare state”, Social Policy and

Administration, 30 (4):324–347.

GREECE: ANTI-POVERTY POLICY 63

Sarris, A. and Zografakis, S. (1999) “A computable general equilibrium assessment of the impact of illegal immigrationon the Greek economy”, Journal of Population Economics, 12:155–182.

Sen, A. (1995) “The political economy of targeting”, in D.van de Walle and K. Nead (eds), Public Spending and thePoor: Theory and Evidence, Baltimore, MD: Johns Hopkins University Press.

Sotiropoulos, D. (2003) “‘Aspect of Babylon’: interpretations of the post-war development of the welfare state inGreece”, in D.Venieris and C.Papatheodorou (eds), Social Policy in Greece: Challenges and Perspectives, Athens:(in Greek), pp. 89–131.

Svoronos, N. (1972) Histoire de la Grèce moderne, Paris: Presses Universitaires de France.Tsakloglou, P. and Mitrakos, T. (forthcoming) “Inequality and poverty in the last quarter of the twentieth century”, in

E.Mossialos and M.Petmesidou (eds), Social Policy Developments in Greece: A Halfway House Reform?,Aldershot: Ashgate.

van Oorschot, W. (1991) “Non-take up of social security benefits in Europe”, Journal of European Social Policy, 1 (1):15–30.

van Oorschot, W. (2002) “Miracle or nightmare? A critical review of Dutch activation policies and their outcomes”,Journal of Social Policy, 31 (3):399–420.

Venieris, D.N. (1996) “Dimensions of social policy in Greece”, South European Society and Politics, 1 (3):260–269.

64 MANOS MATSAGANIS

3Italy—striving uphill but stopping halfway

The troubled journey of the experimental minimum insertion income

Stefano Sacchi and Francesca Bastagli

Introduction

Largely marginal in Italy’s national debate and policy agenda until the 1990s, the issue of poverty andsocial exclusion gained increasing salience in the years spanning the millennium. This occurred at the sametime as new and tighter budget constraints were spurring policy innovation, joining elites’ awareness of thestructural imbalances of the Italian welfare state to open a window of opportunity to reform Italy’s socialassistance sector and to devise a strategy to fight poverty and social exclusion.

A project of gradual re-calibration both within social assistance and between social assistance and socialinsurance (especially pensions) was elaborated, resting on the following broad orientations:

• the universalization of potential access to benefits, linked to the implementation of effective selectivitycriteria and mechanisms in ascertaining beneficiaries’ needs and determining the deserving households;

• the establishment of a safety net;• the promotion of a wider array of decentralized services, with a smoothening of territorial disparities;• a rationalization of existing cash transfers, separating more clearly the anti-poverty function from that of

supporting family responsibilities.

A number of important steps were taken with regard to the implementation of this broad project, with theintroduction of some new policy measures, a thorough reform of the social assistance sector merging withthe Italian side of a new EU-wide social inclusion strategy, and the introduction, albeit in an experimentalfashion only, of the missing building block of the Italian welfare state edifice: a generalized safety netpotentially available to everyone in need.

For various reasons, pre-eminent among which is a sweeping constitu tional reform that in 2001rearranged the allocation of legislative powers between the territorial levels of government, this process ofsocial assistance reform has lately ground to a halt, and the establishment of a nation-wide minimumincome scheme with it.

This chapter analyses the rise and fall of a strategy to fight poverty and social exclusion in Italy, againstthe background of the ancestral features of Italian poverty alleviation policy, on the one hand, and of theproblem pressure exerted by poverty conditions, on the other. The next section describes the historical andinstitutional trajectory of poverty alleviation policy in Italy, from national unification to the Berlusconigovernment, with a focus on the developments of the past decade. The third section provides an overallpicture of the structural imbalances of the Italian welfare state and analyses in depth the profile of poverty inItaly. On such a basis, the fourth section reviews those measures that—explicitly or more often implicitly—

can have a mitigating impact on poverty in Italy and assesses them in terms of their effectiveness inreducing poverty and of their distributional consequences. The outcomes of the anti-poverty measurescurrently in operation in Italy immediately make a minimum income scheme conspicuous by its absence.The section ends with an account of the political vicissitudes that have accompanied the mirage of ageneralized safety net, from the introduction of the Reddito minimo d’inserimento as an experimentalprogramme in 1998 to the present day. The fifth section focuses on the experience of the Reddito minimod’inserimento, describing its features and highlighting the main difficulties that have emerged during theexperimentation, with a view to identifying those black spots that would have to be addressed were a minimumincome scheme ever to be introduced in Italy. The final section concludes, making the case for theintroduction of a safety net as an essential level of provision concerning social rights to be guaranteedacross the entire national territory and open in principle to the whole citizenry.

Poverty alleviation in Italy: a historical perspective

The late development of Italy’s social assistance policies

From as early as 1861—the year of national unification—to very recently, social assistance has been largelyneglected by Italy’s policy-makers.1 For the first two decades after unification, the state was content withbroadly regulating non-public—mainly Catholic—charities, on the basis of an 1862 law extending theregulatory framework already in place in Piedmont to the newly-created country. State non-interference inthe social assistance field was upheld, for different reasons, by all the major political forces: liberals,although worried by the increasing strength of the Catholic charities, remained committed to laissez-faireprinciples; Catholics fiercely opposed any state intervention into what had traditionally been their reserveddomain; and the labour movement too was, in its early stages, opposed to state intervention (Ferrera 1986).

Only in 1890 did the state intervene more in the charitable organizations sector, tightening its controlover religious charities.2 Meanwhile, Italy’s governing elites had taken the Bismarckian road to welfare,principally in order to counteract the challenge posed to their rule by the labour movement. Between 1883and 1919 many social provisions were passed, from compulsory insurance against industrial accidents topension schemes, to compulsory unemployment insurance. Since the enfranchisement of the malepopulation in 1912, however, and after the inclusion of socialists and Catholics in the parliamentary arena,the creation and extension of social rights cannot be described only as a concession on the part of the elites,but rather as the outcome of competitive dynamics aimed at winning the (male) popular vote.

Set up for defensive reasons on the part of the elites in a competitive oligarchy or an outcome of thepolitical process in a near-polyarchic polity, Italy’s welfare state was built on occupational lines. Its originalmodel, with social entitlements only granted to those with a stable attachment to the labour market and theirfamilies, made no allowance for comprehensive public schemes geared towards alleviating poverty amongthose unable to work.

Under the fascist regime (1922–43) the state greatly increased its presence in the social assistance sector,setting up many public institutions while still allowing the Church much room for manoeuvre. Insurance-based family allowances, still among Italy’s most important poverty-allevia-tion measures at present, wereintroduced in the late 1930s for dependent workers. Above all—conforming to the corporatist pattern itimposed on the political economy—the fascist state disaggregated social benefits along occupational, evenmicro-categorical lines, a feature that strongly affected the subsequent development of the Italian welfarestate.

66 ITALY: MINIMUM INSERTION INCOME

For two decades after the end of the Second World War, the Italian welfare state expansion took placealong categorical lines, while social assistance was patchily provided by a plethora of public institutions atthe national and local level, flanked by private and Catholic Church-based charities. As for combatingpoverty, a 1952 law established pension minima and related pension supplements.

Between 1969 and 1978 important reforms were passed, regarding pensions, health care, and also socialassistance. In 1969 a pension reform adopted an earnings-related pay-as-you-go system. At the same time,the pensione sociale (social pension) was introduced, a social assistance measure intended for over-65-year-olds with no income.3 In 1978 the National Health Service was created, granting equal and universal accessto health care to all citizens. As for social assistance, in 1977 responsibilities in such a field were devolvedto the regions and local authorities while the state retained authority over measures of a contributory orinsurance nature. In transferring responsibilities, the 1977 reform did not establish guiding principles orgeneral standards to be respected by the regional and local levels of governments: it simply stated that theseprinciples and standards would be set by a national law that would reform the social assistance sector (thefirst since the legge Crispi of 1890), and by regional laws.

Until very recently (Law 328 of 2000), there was no national law on this matter. This “legislative void”allowed local social assistance measures to develop in an autonomous, unco-ordinated fashion, often on thebasis of pre-existing policies and institutional structures rather than on the basis of specific, “local” needs.This has led to an extremely variegated system, in which different regions and municipalities providedifferent measures and define beneficiaries and criteria of access on the basis of their own standards. Hence,the picture is one of social assistance measures which are locally non-homogeneous and nationally stunted.Little wonder that the burden of alleviating poverty at a national level is mainly carried by social insurancepolicies, and in particular by pension supplements, addressing those whose pension benefit is lower than aset threshold, but still conditional on a 20-year-long contribution record. Widely exploited as povertyalleviation schemes (and as instruments for clientelist exchange) are also invalidity pensions: the insurance-based ones (INPS invalidity pensions) until 1984 and, after tighter regulations for the latter was introducedin that year, the income-tested non-contributory ones (civil invalidity pensions).

Teetering, tottering: the rise and fall of a social inclusion strategy in Italy

One peculiar trait of Italy’s social assistance sector is the lack of an established guaranteed minimumincome. Italian social assistance measures are directed at individuals belonging to a number of “categories”of need such as disability and old age. To qualify for assistance benefits, an individual must belong to a“protected” category in the first instance, and then must pass the means test. Such measures arepredominantly transfer-based: they consist mainly of monetary transfers as opposed to services or in-kindbenefits. In 2002, only 20.5 per cent of public social assistance expenditure was spent on in-kind benefits.The remaining 79.5 per cent was spent on cash transfers. Of the latter amount, about 58 per cent (i.e. 46 percent of public social assistance expenditure) was absorbed by civil invalidity pensions, which therefore takethe lion’s share of Italian social assistance.4

In addition to national social assistance schemes, a variety of policies exists—as mentioned earlier—at asubnational (regional or municipal) level. Local minimum income schemes are a significant example.Although such a scheme does not exist at a national level, in some cases measures aimed at providing thepoor with a non-categorical, means-tested benefit (minimo vitale) have been in force at a local level for along time now. In some situations regional laws regulating the establishment of a minimo vitale wereadopted, in others, individual municipalities chose over the years to introduce such a measure independentlyfrom regional legislation. However, the adoption of minimum income schemes at a local level has occurred

STEFANO SACCHI AND FRANCESCA BASTAGLI 67

in a highly heterogeneous way across the national territory. Municipalities such as Turin (1978), Ancona(1981), Catania (1983) and Milan (1989) introduced a minimo vitale in the years indicated in contrast tocities such as Rome and Bari, where a form of minimum income has never been introduced. This localdifferentiation has led to a system of social assistance in which citizens are entitled to benefits that varyconsiderably and that depend not so much on the situation of need per se but, rather, on the place in whichthe person in need lives.

Things, however, started to change from the mid-1990s onwards, at least from a policy-orientedperspective.5 The issue of poverty and social exclusion came to the fore of the political discourse when, inthis period, the wider issue of overarching welfare state reform started to be debated as a consequence ofchanging labour markets, family structures and demographic trends to meet with the now binding post-Maastricht budgetary constraints. In pursuit of membership of the EMU, the government, led by RomanoProdi, appointed a commission of experts in 1997 to put forward proposals to reform Italy’s entire welfarestate building. The Commission for the Analysis of Macroeconomic Compatibilities of Social Expenditure(Commissione per l’Analisi delle Compatibilità della Spesa Sociale), chaired by the economist Paolo Onofriand hence known as “Onofri Commission”, came up in the same year with a blueprint for a thorough reformof the Italian welfare state, comprised of—inter alia—a project of gradual re-calibration both within socialassistance and between social assistance and social insurance (especially pensions).6 Though most of itsproposals regarding pensions and social shock absorbers reform were of little consequence because of theopposition of trade unions and of the relevant parties within the government’s centre-left parliamentarymajority, proposals for a reform of the social assistance sector were at least partly translated into new policymeasures.

Generally speaking, in its report, the commission set forth a policy view inspired by universalism asregards the potential access to a benefit, coupled with effective selectivity criteria in the determination ofthe deserving households. In the years following 1997 some fundamental reforms were initiated along thisview, that is:

• designing an organic system of social assistance (mainly through the introduction of a general frameworklaw);

• experimenting and realizing new programmes;• introducing a new method of evaluation of the financial situation of potential social assistance

beneficiaries (the ISEE, Indicatore della Situazione Economica Equivalente).

The missing pillar of Italian social assistance, a means-tested minimum income programme, started to be“experimented” in 1998.7 Two important poverty-alleviating measures were also introduced at the end of1998: a transfer to large households and a maternity benefit payable to those mothers who are not entitled tothe insurance-based maternity allowance.8

All these changes occurred against the background of an overall process of devolution of administrativepowers from national to subnational (regional, provincial and local) levels of government, which was takingplace in Italy at that time. The directive functions were reserved for the state, while the functions of regionalplanning were assigned to the regions and those of implementation of services to the local authorities.Framework Law no. 328 of 2000, which reformed the institutional setting of Italian social assistance morethan a century after legge Crispi, was drawn up in accordance with such ongoing decentralization and withthe principles of subsidiarity, while at the same time firmly establishing the national government’scompetence for fixing general guidelines. It outlined a detailed division of labour between the various levelsof government, setting in the social assistance field the powers and responsibilities of the central

68 ITALY: MINIMUM INSERTION INCOME

government, of the regions, the provinces and the municipalities (or groupings of municipalities, called“zone”). In this social policy planning system the central level of government retained the power of policysteering and overall co-ordination, mainly through a series of master plans (the Social Plan, the NationalAction Plan on Employment, the Health Care Plan, the Education Plan) and through more circumscribedplans (on disability, children and teenagers, drug addiction, the elderly). All these plans would containmeasures that address social exclusion, but the most important among them was the triennial National Planof Social Interventions and Services (Piano Nazionale degli Interventi e dei Servizi Sociali), henceforth theSocial Plan. Setting out the priorities for action in the social assistance field, this was the main instrument ofnational policy planning, and the foundation of regional and zonal planning (under the Framework Law,regions and zone must each issue regional and zonal social plans). The first (and only, for reasons that willsoon become clear) Social Plan was issued in April 2001 to cover the period 2001–03.9 It identified fivepolicy priorities: supporting family responsibilities, enhancing children’s rights, fighting poverty,supporting dependent persons (especially the severely disabled) through home help services, and promotingthe inclusion of specific problem groups (immigrants, substance abusers, teenagers).

Substantively, the Framework Law promoted policies in the form of services and in-kind benefits tocomplement monetary transfers, and devised the construction of an integrated system of social services andinterventions based on the principles of universal access, policy integration, vertical integration (betweenlevels of government) and horizontal co-operation (between public authorities, NGOs and the socialpartners). Path-breaking in the Italian context of policies against poverty was the principle of universalaccess, with priority given to individuals and families in economic need, or suffering from physical andpsychical disability; traditionally, such policies have been categorical and highly unfair, with those most inneed easily slipping through the safety net. At the same time, the Framework Law addressed anotherweakness of the Italian social assistance sector: the low financial resources destined for it, establishing thatsocial assistance measures be financed with resources from the National Fund for Social Policies.10

In the same year the Framework Law was issued (i.e. 2000), the Lisbon European Council launched aEuropean strategy against social exclusion. Within this overall strategy, the social inclusion process requiresthat every two years each member state draws up a National Action Plan for social inclusion (Nap/incl),setting out in detail its national strategy to combat poverty and social exclusion.11 Although rather poor inquality and non-committal, Italy’s first Nap/incl—presented in July 2001—was entirely built on the newplanning system in the social assistance sector, to the extent that it probably went too far in its reliance onthe possibility of policy planning being easily translated into implemented policies and virtuous outcomes.12

To sum up, the years spanning the millennium witnessed the attempt to develop a comprehensive,general strategy against poverty and social exclusion, set out in the Framework Law 328 of 2000 and in theSocial Plan issued in 2001. All this was accompanied by the introduction of new, important measures toalleviate poverty.

However, the past few years have witnessed some developments which have seriously jeopardized Italy’sachievement of basic standards of social assistance, similar to those commonly enjoyed by citizens of mostWestern European countries.13

The constitutional reform of October 2001, affecting Title V (Titolo V) of the Constitution—whichestablishes the prerogatives of the regions, the provinces and the municipalities—has modified theallocation of competences between the various levels of government in many policy areas.14 Socialassistance is now an area subject to exclusive regional legislation. True, the setting of essential levels ofprovision concerning social rights to be guaranteed across the whole national territory is still in the hands ofthe state, which furthermore retains the power to substitute for lower levels of government in order toensure that citizens enjoy the essential levels of provision, once established.15 But by depriving the national

STEFANO SACCHI AND FRANCESCA BASTAGLI 69

government of those tools for co-ordination and planning that it had acquired shortly before, after a quarter-century wait, the constitutional reform has at least partially disembowelled the reform of social assistanceembodied in the framework law.16 As a consequence of the new constitutional provisions, the state cannotissue a new Social Plan to replace the one that expired in 2003, and the co-ordination of lower-levelpolicy action into a coherent whole is considerably more difficult than before. Implementation of theFramework Law is haphazardly taking place at the only territorial level now vested with legislative andregulatory capacity in the social assistance sector (except for the setting of the essential levels of provision):the sub-national level. The drawback is that, with no guidance on the part of the national level,implementation is left to the voluntary compliance of regional and local governments, thus perpetuatingprecisely what Framework Law 328 of 2000 intended to counteract: territorial disparities in the enjoymentof social rights, particularly between the North and the South.

Even under the new constitutional provisions, however, there might be more scope for action on the partof the central government than what the centre-right government led by Silvio Berlusconi, voted into officein spring 2001, seemed keen to undertake. Apparently, the government was rather eager to consider Law328 of 2000 as obsolete and in February 2003 it issued a rather vague White Paper on Welfare which is oflittle practical use as a guidance document.17 Prima facie, the state has backed away from the socialassistance sector—and consequently from political action to combat poverty and social exclusion—evenmore than would be congruent with the constitutional reform. More accurately, though, it could be said thatthe Berlusconi government has reoriented priorities within the social assistance field away from an overallstrategy and towards haphazard interventions aimed at the elderly and the families.18 This can be gaugedthrough Italy’s second Nap/incl, presented in July 2003 to comply with the second round of the socialinclusion process. Again, no commitments were made, no targets to reduce poverty were set, and no policymeasures for the period 2003–05 were devised.19

Within this changed context, there is no place for a nation-wide social safety net. Far from beingextended to the whole national territory, the minimum income scheme, which was experimented in selectedmunicipalities between 1998 and 2002 and highlighted as a best practice in the first Nap/incl, has beendiscontinued.20 Such a measure could well be one of those essential levels of provision concerning socialrights the state is entitled to set, and one of the easiest to craft, at least as regards its monetary component.However, the Budget Law for 2004 provided for co-funding on the part of the state if a region decides toadopt a measure called Income of Last Resort (Reddito di ultima istanzd). Although the details of suchmeasure are still unclear, it is remarkable that the central government eschews social policy intervention in amatter, that of essential levels of provision, falling within its exclusive competence and leaves suchintervention to regional voluntarism.

A snapshot of problems: poverty and the Italian welfare state

When the focus of analysis is shifted towards policy outcomes, even a casual glance at poverty and socialprotection figures is sufficient to single out some of the characteristics and peculiarities of the Italianwelfare state.

A comparison between Italian social protection expenditure and that of other EU-15 member states isparticularly telling. Although Italy’s share of social protection expenditure in GDP is not too far from theEU-15 average (25.6 per cent against 27.5 per cent of the EU-15 average in 2001), the allocation of socialexpenditure by function indicates some distinct “anomalies”. Figures 3.1 and 3.2 show that Italyconcentrates most of its welfare effort on old age and survivors (62.3 per cent of total social benefits againstan EU-15 average of 43.0 per cent), devotes proportionally less resources than the EU-15 average on

70 ITALY: MINIMUM INSERTION INCOME

sickness, health care and disability (31.8 per cent against 36.2 per cent), while at the same time displaying amarkedly low allocation of social benefits to all other functions: family/children (4.0 per cent against 8.0per cent), unemployment (1.6 per cent against 6.2 per cent) and housing and social exclusion (0.3 per centagainst 3.6 per cent).21

Figure 3.3 shows the allocation of public social protection expenditure among different types of welfareprogrammes. In 2002, 67.7 per cent of total public expenditure on social protection benefits was spent on socialinsurance type programmes, 25.1 per cent was spent on health and 7.2 per cent on social assistance typeprogrammes.22

The data presented above epitomize two distinct imbalances of the Italian welfare state:

Figure 3.1 Social benefits by group of functions 2001 (% of total social benefits) (source: Eurostat (2004)).

Figure 3.2 “Other” social benefits 2001 of total social benefits) (source: Eurostat (2004)).

STEFANO SACCHI AND FRANCESCA BASTAGLI 71

• towards social insurance type programmes as opposed to social assistance ones;23

• towards the protection against certain risks as opposed to other risks or to the protection of otherconditions of need.

The concentration of resources on social insurance type measures implies the greater protection of somecategories to the detriment of others. Social insurance or contribution-based measures are only accessible toindividuals (or to dependants of individuals) who have worked in the formal labour market and that havedone so continuously and long enough to mature an entitlement to a benefit. Hence, the most distinctivetrait of the Italian welfare state is perhaps the dualistic, almost polarized character of the protection offeredto the core sectors of the labour market, on the one hand, and the more peripheral sectors, on the other. Thiscreates gaps of protection at the bottom for groups such as workers with irregular jobs and elderly peoplewith inadequate or no contributionbased entitlements.

As for the second imbalance, when compared to its European counterparts, the Italian welfare stateprovides disproportionate benefits for old age, invalidity, survivors, temporary or partial unemployment,and shortterm sickness, while it grants little protection to the risks connected with a large family, total lackof work or resources, long-term sickness, need of long-term care, and housing problems.24 For instance, foreach euro spent in Italy on benefits related to housing or social exclusion, � 207 are spent on old age andsurvivors’ pensions, as opposed to an EU average ratio of 1:12.25 Cross-cutting with the first, this secondimbalance creates a highly polarized social distribution which, given the geography of Italy’s labourmarkets, also has a strong territorial component. Consistently with the data on poverty below, the ultimatelosers of such a distribution seem to be large families with unemployed spouses in the South.

Figure 3.3 Italy’s public expenditure on social protection benefits by category 2002 (%) (source: Ministerodell’Economia e delle Finanze (2003)).

72 ITALY: MINIMUM INSERTION INCOME

Poverty in Italy

Studies on poverty in Italy provide data on both relative and absolute poverty.26 At a national level, absolutepoverty has followed a relatively stable trend in recent years (see Table 3.1).27 As for relative poverty, thetrend was stable from 1997 to 2001, with a modest fall in 2002.28 Both relative and absolute poverty aredefined in Italy in terms of consumption expenditure and not of income. The reason for this is the higherreliability attached in Italy to consumption surveys than to self-reporting of personal income.29

As for relative poverty, the Italian National Statistical Institute (Istat) sets the relative poverty thresholdaccording to the International Standard of Poverty Line (ISPL): a family of two is poor when its expenditureon consumption is below or equal to national average per-capita expenditure,

Table 3.1 Headcount ratios (households) for relative and absolute poverty (1997–2002) (%)

1997 1998 1999 2000 2001 2002

Relative poverty 12.0 11.8 11.9 12.3 12.0 11.0

Absolute poverty 4.6 4.5 4.8 4.3 4.2 4.2

Source: Istat (2003a).

estimated through the annual survey on household consumption (sample size is over 27,000 households, morethan 1.2 per thousand of Italian household population).

Set in this way, in 2002 the relative poverty threshold was equal to � 823.5 monthly for a family of twocomponents. On the basis of this threshold and using the Carbonaro equivalence scale (Table 3.2) (whichtakes into account only the number of family members and not also possible special needs), the percentageof poor families over the total number of resident families, or poverty headcount ratio, was equal to 11 percent. The percentage of individuals below the poverty line was 12.4 per cent. In absolute figures, this meansabout 2,456,000 households and 7,140,000 individuals. The average poverty gap ratio among the poor,which expresses in proportional terms how much lower the average expenditure of poor families is than theexpenditure level of the poverty threshold, was equal to 21.4 per cent in 2002 (Table 3.3) .30 The analysis ofdata on poverty in Italy highlights its highly uneven distribution both at a territorial level and betweenfamily typologies.

The territorial imbalance of poverty between the Centre-North and the South is perhaps the mostdistinctive characteristic of poverty in Italy (see Table 3.4). In the southern regions and the islands theincidence of poverty in 2002 was more than three times higher than in the Centre, and about four and a halftimes higher than in the North. Intensity of poverty

Table 3.2 The Carbonaro Equivalence Scale

No. of components Coefficients

1 1.00

2 1.67

3 2.22

4 2.72

5 3.17

6 3.60

7 or more 4.00

STEFANO SACCHI AND FRANCESCA BASTAGLI 73

Table 3.3 Headcount ratio and average poverty gap ratio (relative poverty) (1997–2002)

1997 1998 1999 2000 2001 2002

Headcount ratio (%)

Households 12.0 11.8 11.9 12.3 12.0 11.0

Individuals 13.0 13.0 13.1 13.9 13.6 12.4

Average poverty gap ratio (%)

Households 21.5 22.4 22.9 22.5 21.1 21.4

Source: Istat (2003a).

Table 3.4 Relative poverty by geographic area (2002) (%)

North Centre South Italy

Headcount ratio

Households 5.0 6.7 22.4 11.0

Individuals 5.4 7.9 23.6 12.4

Average poverty gap ratio

Households 19.3 20.0 22.3 21.4

Source: Istat (2003a).

is instead similar across the three main Italian areas, particularly after a fall in the South and a rise in theNorth and the Centre in 2002.

As shown in Table 3.5, while the share of Italian households living in the South amounts to one-third,about two-thirds of poor Italian families reside in this area. The imbalance is reversed in the North: it hostsonly one-fifth of Italy’s poor households as opposed to half of Italy’s total number of households.

Tables 3.6 and 3.7 indicate the characteristics of households that are particularly exposed to the risk ofpoverty. Large households, households with children and households in which the person of reference is notemployed are those that experience the highest poverty risk. At least at the national level, gender does notseem to affect the risk of being poor: the relative risk of being poor for a household with a female person ofreference is only marginally higher than for one headed by a male.31

Throughout the country, poverty hits larger families hardest: 23.4 per cent of households of five or morecomponents live in poverty at a national level, compared to 8.8 per cent of single member households. Theanalysis of the household composition also casts light on other poverty risk factors, namely the presence ofunder-age members (less than 18 years old) and of elderly members.32 Households with at least one under-age member show a higher than average poverty rate of 12.8 per cent, those with at least three a value of 25.9 per cent, while in all geographical areas the relative risk of poverty becomes greater than one with thesecond under-age child in the household. In the southern regions and

Table 3.5 Distribution of poor as opposed to residents by geographic area in 2002 (%)

North Centre South Italy

Poor households 21.9 11.8 66.3 100

Resident households 48.0 19.4 32.6 100

Poor individuals 19.4 12.2 68.4 100

Resident individuals 44.6 19.3 36.1 100

74 ITALY: MINIMUM INSERTION INCOME

North Centre South Italy

Source: Istat (2003a).

Table 3.6 Relative poverty rates (%) by geographic area and family characteristics (2002)

North Centre South Italy

Household size

1 member 4.9 3.7 20.0 8.8

2 members 4.7 7.4 24.0 10.7

3 members 3.9 5.8 19.5 8.9

4 members 5.7 8.0 21.1 12.5

5 or more members 11.6 15.0 32.4 23.4

Households with under-age children

1 under-age child 4.2 6.0 18.3 9.2

2 under-age children 7.6 9.8 23.2 15.2

3 or more under-age children 16.7 * 32.9 25.9

At least one under-age child 6.1 7.0 22.1 12.8

Households with elderly

1 elderly 7.3 6.8 27.2 13.4

2 or more elderly 8.0 13.6 33.3 17.4

At least one elderly 7.5 9.1 29.2 14.7

Family type

Lone person <65 years 1.7 * 8.9 3.1

Lone person with or over 65 years 7.7 6.7 26.4 13.3

Couple with head of household <65 years 1.8 * 12.7 4.8

Couple with head of household equal to or over 65 years 7.3 10.9 32.5 15.7

Couple with 1 child 3.5 4.8 18.0 8.1

Couple with 2 children 5.4 8.2 20.2 12.2

Couple with 3 or more children 13.0 11.7 31.8 24.4

Single parent 6.0 7.1 21.4 11.5

Other types 7.3 11.4 35.0 15.7

Overall 5.0 6.7 22.4 11.0

Source: Istat (2003a).Note* Not significant data.

Table 3.7 Relative poverty rates (%) by geographic area and professional situation of the person of reference (2002)

North Centre South Italy

Professional situation of person of reference

Employee 3.4 5.3 17.6 8.5

Self-employed 3.0 2.6 15.0 6.4

Unemployed * * 40.7 32.2

STEFANO SACCHI AND FRANCESCA BASTAGLI 75

North Centre South Italy

Retired from work 6.2 8.7 26.2 12.3

Overall 5.0 6.7 22.4 11.0

Source: Istat (2003a).Note* Data not significant due to low number.

the islands, the incidence of poverty for households with at least three under-age children amounts to 32.9per cent. As for the elderly, households with at least one member over 65 display a higher than averagepoverty rate, namely 14.7 per cent. In the South, one in three households with two or more members over 65is poor.

Looking at single-member families, the incidence of poverty among over-65s who live alone is four timeshigher than among singles aged less than 65 (13.3 per cent as opposed to 3.1 per cent). By the same token,couples where the head of household is over 65 are over three times more at risk of being poor than coupleswith a head of household aged less than 65 (poverty rates are 15.7 per cent and 4.8 per cent, respectively).Nonetheless, a comparison between couples where the head of household is over 65 (poverty rate: 15.7 percent) and couples with three or more children (poverty rate: 24.4 per cent) shows that the outcome of theItalian welfare state is more favourable to the elderly than to large families. It should also be noted that inItaly the poverty rate among single parent households is not significantly higher than average.

When the professional situation of the person of reference of the household is considered, the highestincidence of poverty occurs for those where the person of reference is unemployed: 32.2 per cent in 2002,almost three times higher than the overall poverty rate. The relative risk of poverty is higher than one alsofor those households where the head is retired (see Table 3.7).

As for absolute poverty, the Istat survey defines it in terms of a basket of essential, basic goods andservices necessary to guarantee a socially acceptable living standard. The monetary value of the basket ofgoods and services is comprised of a food component, a housing component, a component of depreciationof basic durable goods (refrigerator, washing machine, television) and a residual component that takes othercosts (transport, clothing, recreational activities) into account. The basket does not include health costs oreducation costs, since both health and education are assumed to be obtained free of charge through theNational Health Service and the national education system respectively. In 2002, the absolute povertythreshold was set at � 573.6 (monthly) for a family of two. For a single individual the threshold was equal to� 382, higher than the threshold that entitled an individual to the minimum insertion income.

The trend followed by absolute poverty between 1997 and 2002 looks rather stable (see Table 3.8).Households below the absolute poverty threshold in 2002 made up 4.2 per cent of Italian households, 926,000 units in absolute terms. This meant 2,916,000 individuals, 5.1 per cent of Italy’s population. Likerelative poverty, also absolute poverty has higher incidence in the regions of the South and isdisproportionately concentrated there (Table 3.9).

Absolute poverty is highest among large households: those with four or

Table 3.8 Absolute poverty (headcount ratio and average poverty gap ratio) (1997–2002) (%)

1997 1998 1999 2000 2001 2002

Poverty rate Households 4.6 4.5 4.8 4.3 4.2 4.2

Intensity of poverty Households 18.6 20.0 19.6 19.3 19.3 19.6

Sources: 1997–2000 Commissione d’Indagine sull’Esclusione Sociale (2002); 2001–02 Istat (2003a).

76 ITALY: MINIMUM INSERTION INCOME

Table 3.9 Incidence and distribution of absolute poverty by geographical area (2002) (%)

North Centre South Italy

Headcount ratio 1.7 2.7 8.9 4.2

Poor households 19.8 10.2 70.0 100

Resident households 48.0 19.4 32.6 100

Source: Istat (2003a).

more members display a poverty rate of 7.0 per cent (see Table 3.10). Family types more at risk of absolutepoverty coincide with those of relative poverty. Conditions of extreme deprivation are observed amongcouples with at least three children, with an incidence of absolute poverty in 2002 as high as 14.4 per cent,more than three times higher than the national average. The relative risk of absolute poverty is alsosignificantly higher than one for singles aged 65 and more.

Figure 3.4 At-risk-of-poverty rate after social transfers, EU-15 2000 (ECHP 2001) (source: Eurostat, StructuralIndicators, http://www.europa.eu.int/comm/ eurostat).

In a comparative perspective, Italy ranks among the six EU-15 countries with the highest incidence ofpoverty (Figure 3.4) and the highest rate of persistence in poverty (Figure 3.5).33

Comparing relative risks of poverty of selected categories in Italy and the EU may be more interestingthan simply comparing overall poverty rates. Data collected by Boeri and Perotti (2002) show that—in 1995—in the Italian sub-sample of European Community Household Panel (ECHP) the risk of poverty ofhouseholds with children relative to that of households without children was markedly higher than in theoverall ECHP sample (Table 3.11).34 The situation becomes even worse when the focus is shifted frompoverty in a single year to persistence in poverty: compared to households without children, in 1995households with children were twice as likely to have been poor for three years in a row in the Italian as inthe overall ECHP sample. The same obtains when two-adult households are considered. But the moststriking findings, in a comparative perspective, are those shown in the two bottom rows of Table 3.11: whilein ECHP countries, as a whole, one-adult households, with or without children, are

STEFANO SACCHI AND FRANCESCA BASTAGLI 77

Table 3.10 Absolute poverty incidence by family size and type (2002)

Family size and type (%)

Household size

1 member 3.6

2 members 2.8

3 members 2.9

4 or more members 7.0

Family type

Single person below 65 years 1.2

Single person of 65 years or more 5.6

Couple with head of household below 65 years 0.8

Couple with head of household aged 65 years or more 4.4

Couple with one child 2.6

Couple with two children 4.5

Couple with at least three children 14.4

Single parent 3.4

Other 7.2

Overall 4.2

Source: Istat (2003a).

considerably more at risk of being poor than two-adult households with children (as much as one and ahalf times more in the case of a single adult with children), the pattern for Italy is reversed.

Figure 3.5 At-persistent-risk-of-poverty rate after social transfers, EU-15 2000 (ECHP 2001) (source: Eurostat,Structural Indicators, http://www.europa. eu.int/comm/eurostat).

78 ITALY: MINIMUM INSERTION INCOME

Another comparatively peculiar trait of Italy’s poverty profile regards the remarkable difference inrelative risks of poverty for households

Table 3.11 Risk of poverty for households with children relative to that in other selected family types (ratios), 1995 and1993–95 for data on persistence

Italy ECHP countries

With children/without children 1.511 1.161

Persistence 2.605 1.382

Two adults with children/Two adults without children 1.985 1.349

Persistence 3.556 1.707

Two adults with children/One adult without children 1.036 0.789

Two adults with children/One adult with children 1.179 0.627

Source: Rows 1–2: table 2.1 in Boeri and Perotti (2002, 49–50); rows 3–6: own elaboration on data whereof.

headed by an over-65-year-old depending on whether or not at least one family member receives a pension.Elaborating on data provided by Boeri and Perotti, we find that the risk of poverty for a household headedby an over-65-year-old where no member receives a pension relative to that for a same-type household withat least one pension is more than one and a half times higher in Italy than in ECHP countries as a whole (2.49 as opposed to 1.55).35 This means that, for households headed by an elderly person, the cost (in terms ofpoverty risk) of not receiving pension benefits is far higher in Italy than in ECHP countries as a whole.

To summarize, findings all consistently point in the same direction: Italy’s poverty is concentrated in theSouth, in large households, in families with three or more children, in households headed by anunemployed and among elderly people with no entitlement to a contributory pension. This immediatelyleads to poverty alleviation measures: their design, their effectiveness, their distributional consequences.

The picture today

Italy’s poverty alleviation policies

Identifying national welfare policies with an impact on poverty is no easy task in Italy. Different studieshave adopted different criteria of definition and have singled out different policies. This section lists onlythe principal national social protection measures which reach the poor and distinguishes social insurance-type measures from social assistance-type ones principally on the basis of the underpinning logic: anindividual’s entitlement to measures of the former type depends primarily on his/her employment history.The only measure of a universalistic nature among Italy’s cash transfers is the Care allowance (indennità diaccompagnamento), aimed at individuals unable to ambulate or to care for themselves and thus in need ofcontinuous assistance, and who have not been hospitalized in free-of-charge public institutions for morethan a month.36 This benefit is conditional on medical tests on the physical condition of claimants, but is notmeans tested. Over the years, given demographic trends, it has become more and more important as asource of social protection for the frail elderly.

Among the social insurance type measures, of the utmost importance are the following:

• Family allowance (assegno per il nucleo familiare). This is an income-tested monthly transfer reservedto households of dependent workers or retired ex-dependent workers with family burdens.37 The amount

STEFANO SACCHI AND FRANCESCA BASTAGLI 79

of the transfer is positively correlated with the dimension of the household and negatively correlated withits income. To give an example, for the period July 2003–June 2004, a household with both parents andat least one under-age child, with no disabled members and a yearly income below � 11,700, received amonthly transfer (for thirteen months) of � 250 if comprised of four members, of � 359 if comprised offive.

• Pensions supplements (integrazione al trattamento minimo delle pensioni). This is a benefit granted tobeneficiaries of an earnings-related pension aimed at topping their pension benefit up to a statutoryminimum, about � 5,227 per year in 2003 (about 23 per cent of per capita GDP in that year). The receiptof the supplement is conditional on a test on the taxable income of the potential beneficiary (plus, forthose who claimed this benefit after 1994, that of the spouse) and to contribution requisites (twenty yearsof contributions) .38 Over-60s are entitled to higher benefits (maggiorazioni sociali) that vary with age.In particular, the Budget Law for 2002 raised the benefit for over-70s, which in 2003 were entitled to ayearly top-up of � 6,836 when receiving an income (defined in a wider way than taxable income) lowerthan such an amount, if not married.39 Individuals who have entered the labour market after 31December 1995 will not be entitled to this benefit, since with the shift to a contribution-based pay-as-you-go system their pension benefit will be a fraction of accrued contributions, with no supplements.Instead, they might be entitled to the Social Allowance (see below).

• Invalidity pensions (pensione di inabilità and assegno ordinario di invalidità). These benefits are paid toworkers with at least five years of contributions (three of which in the past five years). Eligibility isconditional on a medical test. The assegno ordinario di invalidità can be combined with labour income,varies with income and is granted against a permanent reduction of work capacity of at least two-thirds.40 The pensione di inabilità is a fully-fledged pension, granted to individuals meeting theaforementioned contribution requirements and completely unable to work.41 Until a reform in 1984tightened medical criteria and introduced periodic reviews of the physical conditions of the beneficiary,these programmes served as a functional equivalent of a missing safety net, particularly in the South. Bythe same token, they were hard currency for clientelist exchanges between politicians and beneficiaries/voters (Ferrera 1996).

Among the social assistance-type measures, the following are to be high-lighted:

• Civil invalidity pensions (pensione di inabilità civile and assegno di assistenza). These are non-contributory schemes, therefore, they can be claimed by those who do not meet the contributionrequirements necessary for insurance-based invalidity pensions. In 2003, the pensione di inabilità civilewas reserved for those totally unable to work, earning less than � 13,100 per year, while the assegno diassistenza is aimed at working-age individuals with a working capacity reduced by at least 74 per centand a yearly income lower than � 3,850, and it is replaced by social pension when the beneficiary turns 65years of age. For both measures, the benefit for 2003 was about � 224 per month (thirteen months), about13 per cent of per capita GDP on a yearly basis, and the income test is strictly individual, regardless ofthe size of the family the beneficiary belongs to.42 After 1984, these programmes replaced the INPSinvalidity pensions as core instruments in the clientelistic mechanism outlined above.

• Social pension (pensione sociale). This is a minimum income scheme for over-65-year-olds who are notentitled to an earnings-related pension, and thus not to pensions supplements. For those recording zeroincome, in 2003 the benefit amounted to � 3,846 per year: about 17 per cent of per capita GDP. Thebenefit is subject to an income test: for an unmarried claimant the social pension tops personal income upto the level of � 3,846; for a married claimant it is granted —also within the limit of � 3,846—to complement

80 ITALY: MINIMUM INSERTION INCOME

the couple’s joint income up to the ceiling of � 13,253.43 Higher benefits introduced for over-70-year-oldsby the Budget Law for 2002 apply. Since 1996, new claimants are not eligible for this measure, and haveaccess to the social allowance instead. The social pension will continue to benefit elderly who submittedthe claim before 31 December 1995.

• Social allowance (assegno sociale). Introduced in 1995, this is an income-tested benefit geared towardsover-65-year-olds who, for lack of contribution seniority, are neither eligible to an earnings-relatedpension nor to a contribution-based one, or who in the contribution-based regime introduced in 1995have matured the entitlement to a benefit lower than the social allowance itself. It has replaced the socialpension for new claimants since 1996. For those with no income, in 2003 the benefit amounted to � 4,667per year: about 20 per cent of per capita GDP. The social allowance is subject to an incomes test (relevantincome is defined in a wider way than taxable income); it is granted to complement personal income upto a � 4,667 limit if the claimant is unmarried, and to complement the couple’s joint income up to a � 9,334 ceiling, but only within the � 4,667 limit, if the claimant is married.44 Higher benefits introduced forover-70-year-olds by the Budget Law for 2002 apply.

• Benefit to families with more than three children (assegno per il nucleo familiare con almeno 3 figliminori) or benefit to large families. This is aimed at alleviating poverty in large households, exactlythose that experience a high poverty risk as the data examined earlier show. Italian and EU citizenfamilies with three or more children (under the age of 18) can claim this measure, introduced by theBudget Law for 1999. The monthly cash transfer amounted to � 113 for a family of five components in2003 and is paid for thirteen months with the possibility of being renewed. The measure is means tested;the method used is the Indicatore della Situazione Economica Equivalente (Indicator of EquivalentEconomic Situation), ISEE, introduced in 1998. For a family of five, the ISEE threshold for this measurewas set for 2003 at � 20,381. In 2000, 244,000 families (1.1 per cent of Italian families) benefited from thismeasure (Commissione d’Indagine sull’Esclusione Sociale 2002). They were mainly concentrated in theMezzogiorno.45

• Maternity cheque (assegno di maternità). Also introduced by the Budget Law for 1999, it can be claimedby mothers who are not entitled to the insurance-based maternity allowance. Italian citizens, EU citizensand third-country nationals in possession of a residence permit can claim this benefit. The event givingrise to the entitlement is not only the birth of a child, but also pre-adoption custody and adoption of anunder-age child. The benefit amounted to � 271 monthly per newborn child in 2003 and is paid for fivemonths. The measure is means tested through the ISEE; for a family of two parents and one child theISEE threshold was set in 2003 at � 28,308. In 2000, about 173,000 mothers benefited from the maternitycheque.

(ibid.)It seems clear that this list of policy interventions is mainly aimed at the elderly and the disabled (or better,those unable to work). Benassi (2000) estimates that, not taking into account the most recent measures(benefit to large families and the maternity cheque), more than 90 per cent of total expenditure on anti-poverty measures in Italy (all measures with an anti-poverty function, not only those listed here becausethey are deemed the most important) goes to these two categories. Moreover, and still disregarding the twonewest measures, a striking feature of the schemes listed here is that they do not take into account familysize, nor family members’ condition (disabled members, for instance). Civil invalidity pensions makereference to the claimant’s income only, while pensions supplements, social pension and social allowancetake into account the income of both the claimant and the spouse in a simple manner, without applying an

STEFANO SACCHI AND FRANCESCA BASTAGLI 81

equivalence scale. As for the family allowance, according to the Commission of Inquiry on SocialExclusion, “the amount of the benefit varies from household to household, from income bracket to incomebracket with no relation to any equivalence scale whatsoever, to the extent that it is impossible to define in aprecise way the logical framework of the scheme” (Commissione d’Indagine sull’Esclusione Sociale 2003:35). Finally, the income tests take place with income definitions that vary across the different measures.Only the more recent measures, that is benefit to large families and the maternity cheque, take advantage ofan equivalence scale (the ISEE equivalence scale) and of a rigorous instrument to ascertain the claimant’seconomic conditions, i.e. the ISEE.

The Indicator of Equivalent Economic Condition (Indicatore della Situazione Economica Equivalente)(ISEE), set up in an experimental fashion in 1998 and then finally regulated in 2000, is a method to assessthe economic situation of those claiming “social benefits or services which are not destined to the generalcitizenry or however linked in their amount or in their cost with given economic situations” (LegislativeDecree no. 109 of 1998). The economic situation of the claimant is defined in terms of both income andwealth, adding to the income component a wealth component.46 The unit of reference is the household, sothe total components of each member are summed up and an equivalence scale is applied. The list ofincome sources and types of assets to be considered is quite comprehensive. As for the income component,it is made up of the personal income tax base (adjusted when the claimant is self-employed) and of aconventional financial income, calculated applying to the stock of financial assets the average yearly yieldon ten-year Treasury bonds. A deduction for rental costs of tenants applies, up to � 5,165. As for the wealthcomponent, it is given by the sum of both real estate and financial assets, multiplied by a coefficient of 0.2.Substantial disregards apply: � 15,494 for financial wealth and up to � 51,646 if the household owns itshouse. The two components are summed up to give the ISE (Indicatore della situazione economica,Indicator of Economic Condition). An equivalence scale is then applied to take into account both differentfamily sizes and special needs within the family.47 The indicator obtained by applying the equivalence scaleto the ISE is the Indicatore della Situazione Economica Equivalente, ISEE.

So far, at the national level, the only social protection measures to have adopted the ISEE to determineaccess are those introduced by Budget Law for 1999: the benefit to families with more than three childrenand the maternity cheque. Although its experimental phase is now over, ISEE has never been applied to theother poverty alleviation measures listed in this section. The reason why is straightforward. If applied on ageneralized basis to all means-tested schemes, the ISEE would undoubtedly modify the current pool ofbeneficiaries, extending access to citizens who are now excluded from the benefits and denying access tosome of the current beneficiaries (for instance, those who own conspicuous assets despite reporting lowincomes).48 Much less politically risky then appears the application of the ISEE to brand new programmes,as was done with the measures introduced in 1999.

The impact of Italy’s poverty alleviation policies

A review of the main policy measures that, explicitly or more often implicitly, in the absence of ageneralized safety net still provide some protection against the risk of being poor cannot disregard theirdistributional features (whom do they benefit?) and their effectiveness (do they manage to reduce poverty?).

As for distributional outcomes, using a tax-benefit microsimulation model and data from the 1998 Bankof Italy sample survey of household income and wealth, Baldini et al. (2000) estimate how different incomedistribution deciles are reached by the expenditure on the various meas-ures.49 Among the most strikingresults, it can be found that almost a third of expenditure on pensions supplements goes to the richest half ofthe population, and that the same is the case for social pensions (see Table 3.12). Even more strikingly,

82 ITALY: MINIMUM INSERTION INCOME

more than 15 per cent of expenditure on pensions supplements goes to the richest 30 per cent of thepopulation (more than 10 per cent in the case of the social pension). As for the civil invalidity pensions,more than 30 per cent of expenditure goes to the poorest

Table 3.12 Percentage of expenditure on selected poverty alleviation measures going to various richest populations

Richest % of thepopulation

Family allowance Pensionssupplements

Social pensions Civil invaliditypensions

Invalidity pensions

Whole population 100.00 100.00 100.00 100.00 100.00

Richest 90 77.21 82.18 65.84 68.23 63.12

Richest 80 49.80 62.17 55.72 61.96 51.57

Richest 70 26.45 50.36 44.86 54.52 41.43

Richest 60 15.02 41.68 34.69 34.13 29.13

Richest 50 8.44 30.85 28.17 27.30 23.16

Richest 40 3.50 23.28 18.34 23.75 18.57

Richest 30 1.01 15.36 10.94 11.55 9.32

Richest 20 0.26 9.41 8.04 4.81 7.38

Richest 10 0.07 5.16 4.79 1.39 4.75

Source: Authors’ own elaboration on data in Table 1 in Baldini et al. (2000, p. 7).NoteReference variable is disposable equivalent income.

(first decile), but almost 30 per cent goes to individuals who surely are not poor and more than 10 per centto the richest 30 per cent of the population, despite the fact that this measure is means tested (although witha rather high threshold). The anti-poverty function of insurance-based invalidity pensions is neater: morethat 35 per cent of expenditure goes to the first income decile, almost 50 per cent to the first two deciles.

The ability of reviewed measures to impact on poverty is assessed by Toso (2000b) who, using the samemethodology as Baldini et al. (2000) and data from the 1995 Bank of Italy survey, shows their marginalimpact on poverty incidence, that is the extent to which each measure reduces the poverty headcount ratiowhen it is introduced after all the other measures. The measure which seems to have the highest marginalimpact on poverty is the pensions supplements, that with the lowest impact is the civil invalidity pensions.Such results are consistent with the last column in Table 3.13, taken from Baldini et al. (2000): among thereviewed measures, pensions supplements is the most effective in reducing the gap between the averageincome of the poor and the poverty line (it fills almost 18 per cent of such gap), while civil invaliditypensions is the least effective measure (less than 2 per cent).50 Despite the massive resources absorbed bythe latter measure (almost half of total public expenditure for social assistance benefits in Italy), itseffectiveness in contrasting poverty appears to be very limited indeed.

What we have just said about the effectiveness of pensions supplements must not seem to contradict whatwe observed earlier regarding their low redistributive capacity; a measure may well be effective uniquely bydint of the massive financial resources destined to it. This seems to be precisely

Table 3.13 Targeting indicators of anti-poverty measures in Italy

% of expenditure reaching the poor % of the poverty gap filled by the measure

Family allowance 58.60 11.37

STEFANO SACCHI AND FRANCESCA BASTAGLI 83

% of expenditure reaching the poor % of the poverty gap filled by the measure

Pensions supplements 43.79 17.88

Social pensions 51.96 4.99

Civil invalidity pensions 42.84 1.84

Insurance-based invalidity pensions 54.58 7.57

Benefit for large families 96.44 1.18

Maternity cheque 48.48 0.32

Minimum income scheme 94.81 8.99

Source: Tables 2 and 7 in Baldini et al. (2000, pp. 8 and 18).

the case of pensions supplements: the first column in Table 3.13 reports, for each measure, the share ofexpenditure on each measure reaching households which are poor before the transfer. Thus, more than 56per cent of expenditure on pensions supplements goes to households that are not poor before receiving thetransfer. Even worse is the target efficiency of civil invalidity pensions. That resources poured onto a policymeasure do intervene between its target efficiency (its capacity to reach only the poor) and its effectivenessin reducing poverty can be observed by looking at the benefit for large families, with high target efficiencybut scanty resources to make a major impact on poverty. It is also worth noticing that, although this is onlya simulation exercise, a minimum income scheme moulded on the monetary component of the Redditominimo di inserimento (see below) would be both efficient in reaching only the poor and effective inreducing poverty, and at the same time much less costly than civil invalidity pensions.51

Finally, it must be noticed how the ever increasing use of tax instruments in Italy (tax deductionsreducing taxable income and tax credit allowances reducing tax liability) typically has negligible effects onpoverty. This occurs because introducing, say, tax credit allowances can be relevant if the target individualhas some taxable income and therefore some tax liability, but is not relevant at all when he or she has noincome, as is often the case with the poor. This can be gauged by looking at the children’s allowanceintroduced by Budget Law for 2002, which raised the tax credit to � 516 per each dependent child forhouseholds with a total income lower than � 51,646. The public research centre reporting to the Parliamenton the forecasted effects of budget laws, ISAE, in 2001 forecast that this measure would have very little, ifany, impact on poverty (ISAE 2001).52

This argument seems to be relevant also for the “no tax area” introduced by Budget Law for 2003.53 Withthe obvious exception of a negative income tax, in case of no earned income, the tax instrument has noconsequence whatsoever. In this case, social policy measures are needed. An example of such measures isthe benefit for each second child (Assegno per ogni secondo figlio) introduced in 2003: Italian or EC-country national women who reside in Italy receive a cash transfer of � 1,000 for any child following theirfirst one, whom they give birth to or adopt between 1 December 2003 and 31 December 2004. The cost ofthe benefit is estimated at � 285,000,000 for the whole period (Istat 2003b). It is well known that the purposeof this universal measure—which is flat rate but not means tested—is to enhance births. Given the paucity ofthe benefit, it is highly unlikely that it may reach its pro-natalist goal. However, on an equal footing as manyof the measures reviewed here, it may well have some indirect effect on poverty, being a transfer to thosebelow the poverty line who would have a child anyway, irrespective of the benefit becoming available.According to the National Bureau of Statistics (Istat), however, the impact on poverty would be modest,since the measure would make the poor slightly better off without being able, on average, to make them riseabove the poverty line.

84 ITALY: MINIMUM INSERTION INCOME

From the RMI to the RUI: the delusion of a minimum income scheme in Italy

As clearly highlighted in the last section, any review of poverty alleviation policies in Italy immediatelymakes a non-categorical measure of income support conspicuous by its absence. A minimum income schemeaccessible to every citizen or even resident, irrespective of their age or physical condition, is in place in allEU-15 member states, except for Greece and Italy. Such a scheme started to be “experimented” in 1998 in39 Italian— mostly southern—municipalities under the name of Minimum Insertion Income (RedditoMinimo di Inserimento, RMI).54

The RMI was a measure directly aimed at alleviating poverty and social exclusion and consisted of amonetary component and an “activation” component: a potential beneficiary’s entitlement to the monetarycomponent was conditional on participation in programmes of insertion built on his/her specific situation ofneed or exclusion. For a single member household, the RMI amounted to the difference between anestablished threshold (� 274 in 2001, about 14.6 per cent of per capita GDP on a yearly basis) and themonthly disposable income of the beneficiary, while for different household sizes the threshold wasadjusted according to the ISEE equivalence scale. The programmes of reinsertion were personalized, aimedto recuperate and promote personal abilities and to rebuild social networks, and had to involve all the membersof the household. The influence of the French Revenu minimum d’insertion is evident.

The experimentation of the RMI was initially established for a period of two years, with the deadline setat 31 December 2000. Between 1998 and 2000, the RMI experiment involved over 34,000 families and theexpenditure on benefits amounted to � 220,000,000 approximately.

For the first time ever in Italy, evaluation was carried out on a welfare measure. An Evaluation Report onthe experience of the first two years of RMI experimentation, that was provided by the institutive law andcommissioned by the Welfare Ministry, was compiled in June 2001 by an association of independentresearch centres.55

The Decree instituting the experimental measure in 1998 provided that by mid-2001 the governmentshould report to the Parliament on the experimentation, while Law 328 of 2000 formally instituted the RMIas a general measure in the fight against poverty, bound to absorb other benefits such as the social pensionand social allowance, but left it to a subsequent law to define the details of the programme and, moreimportant, its launch as a fully-fledged measure. Without waiting for this to happen, the Budget Law for2001 (passed in December 2000) determined the extension of the experimentation of the RMI by two years,until 31 December 2002, and to include 306 municipalities, primarily in the South.56

This extension has proved to be fatal for the RMI. In May 2001 the centre-right coalition led by SilvioBerlusconi won the general election. Italy’s Nap/incl 2001–03, issued in July 2001 under the Berlusconigovernment but politically ascribable to the former centre-left government, still highlighted the RMI as abest practice, but this was the last time an official document of the government praised the scheme.

The evaluation report was never discussed by the government in Parliament, but the problems, laid barein the report, that local authorities encountered in implementing and administering the RMI were picked upby the Berlusconi government as the chance to get rid of the scheme, considered a leftover of the centre-leftgovernments.

The “Pact for Italy” (Patto per l’Italia), an agreement signed in July 2002 between the government andtwo of the three main Italian trade unions, tolled the bell for the RMI.57 The Pact stated that “theexperimentation of the RMI led to ascertain the impossibility to identify by means of a state bill those whohave a right to such safety net” and expressed the signatories’ preference for co-funding on the part of thestate of minimum income programmes at the regional level. Under the heading “income maintenance of lastresort”, it affirmed that “the income maintenance system will be completed by a measure of last resort,characterized by solidaristic elements and financed by the general revenue”.58

STEFANO SACCHI AND FRANCESCA BASTAGLI 85

The RMI experimentation, set to expire on 31 December 2002, was not mentioned by the Budget Law for2003, while the White Paper on Welfare, issued in February 2003, restated the negative assessment on theRMI and maintained that “by the year 2003, with the expiration of the RMI experimentation, the incomemaintenance system is completed by a measure of last resort”, called Income of Last Resort (Reddito diultima istanza, RUI) (Ministero del Lavoro e delle Politiche Sociali 2003a:37).

At the same time, due to delays in the transfer of funds from the central government to the municipalitiesinvolved in the experimentation, the RMI expiry date was postponed to 31 December 2004, with noadditional funds but for 19 municipalities, to which extra funding was allocated so as to allow them toprovide “an exceptional supply of benefits” until June 2003.59

The second Italy’s National Action Plan for Social Inclusion, presented in July 2003 to cover the period2003–05, stated again the government’s intention to introduce the Reddito di ultima istanza. It alsomaintained that:

the experience of Minimum Insertion Income highlighted…some problems, partly attributable to thefeatures of the income support component, partly to the scarce capacity [on the part of the localauthorities] of design and implementation of the social insertion component, and partly to thefunctional overload caused by long-standing faults of Italy’s welfare system.

(Ministero del Lavoro e delle Politiche Sociali 2003b:27–28)

This is not, in our view, an inaccurate statement at all: the RMI experience in the experimentation stageactually highlighted many pitfalls of the scheme in both its monetary and insertion components, in both theprogramme design and its implementation. The next section will aim to cast light on these problems. Butthis is precisely what the experimentation should have been useful in doing: to identify black spots and jamsthat should have led to a thorough, even painful if necessary, programme overhaul before its extension tothe whole territory. But the baby was thrown out with the bath water, mainly for ideological reasons (to getrid of anything done by the former centre-left governments).

Finally, the Budget Law for 2004 provided for the introduction of the RUI. It stated that:

the state contributes to financing the regions which establish the Income of Last Resort as a measureof economic accompanying to social reinsertion programmes, aimed at households at risk of exclusionand whose members are not beneficiaries of social buffers for those without work.60

The phrasing of the provision is highly ambiguous, while the Budget Law refers to subsequent regulationthat would set the details for implementation, but that was not yet issued at the moment of writing thischapter (May 2004). The main source of ambiguity relates to the relationship between the state and theregions in the establishment of the measure: will all the regions have to implement the RUI, or will it be thechoice of each region whether or not to introduce a minimum income scheme? (Unless the RUI is listedamong the essential levels of provision concerning social rights to be guaranteed across the whole nationalterritory, something which is out of the question for the time being, it is difficult to foresee how the state canforce regions to legislate on matters falling within their exclusive competence.) This and other related issueswill be dealt with in the Conclusion. Whichever shape the RUI may take, however, it is to be hoped that itsdesign will take into account the lessons that have emerged from the experimentation of the RMI.61

Moreover, such lessons may well be of interest to countries other than Italy, where institutional capabilitiesare similarly low. It is therefore the experience of the RMI experimentation that we turn to in the nextsection.

86 ITALY: MINIMUM INSERTION INCOME

Italy’s minimum insertion income (RMI)

In December 1997, the Prodi government’s Budget Law for 1998 addressed the proposal put forward by theOnofri Commission, regarding the introduction of a minimum income scheme and in June of the followingyear, the Reddito Minimo di Inserimento (Minimum Insertion Income), was introduced in thirty-nine Italianmunicipalities. In the words of the institutive decree (Legislative Decree 237 of 1998), the RMI was ascheme designed to act as a non-categorical guaranteed safety net to “counter poverty and social exclusion,through the support of economic and social conditions of persons exposed to the risk of socialmarginalization and unable to support themselves and their children for psychical, physical or socialreasons”. The scheme consisted of two elements: a cash transfer and an insertion programme. The RMIpursued “the social integration and economic independence of recipient individuals and households throughpersonalized programmes and monetary transfers that integrate income”. The thirty-nine RMImunicipalities were selected on the basis of socioeconomic indicators, initially for a period of two years,beginning in December 1998 and ending on 31 December 2000.62 Five of the thirty-nine municipalitieswere chosen in the north of the country, ten in the centre and twenty-four in the south. Figure 3.6 shows themunicipalities involved.

In 2000, the RMI experimentation was extended by two years, until 31 December 2002 and to include atotal of 306 municipalities. The new RMI municipalities were those belonging to the territorial pacts ofwhich the initial 39 municipalities were a member.63 The official reason behind involving all themunicipalities linked through a territorial pact to an RMI municipality was connected with the hypothesisthat the insertion programmes are more effective where there are developed networks of territorialrelationships among the offices in charge of administering the programme, other local public institutions,the social partners and non-profit organizations.64

Although legislation set the final deadline for the experimentation at 31 December 2002, the RMI wasbeing administered after that. In April 2003, nineteen of the thirty-nine original RMI municipalities receivedfurther funding to continue the payment of benefits until June 2003. The government also agreed that all themunicipalities that had sufficient funds could continue the experimentation until the end of 2004.

This section provides a description and an analysis of the first phase of the RMI experience, drawingprimarily on the Evaluation Report completed by Irs, Cles and Fondazione Zancan and aims to highlightaspects that are critical in the design and implementation of an effective social safety net in Italy.65

Access to the RMI and criteria of eligibility

The RMI could be claimed by Italian or EU citizens who had resided in an RMI municipality for at leasttwelve months and to third-country nationals or stateless individuals who had resided in an RMImunicipality for at least three years. For working-age beneficiaries not employed and able to work, it wascompulsory to agree to attend vocational training or accept work offers. The monthly disposable income ofa potential individual beneficiary had to be equal to or below the RMI threshold set at � 258 in 1999, and at� 268 in 2000 (� 274 in 2001). The ISEE equivalence scale was used to adjust the threshold for families oftwo or more members.66 Although the household was the unit of reference for the RMI and the ISEEequivalence scale applied, the ISEE itself was not used in the process of assessing a claimant’s income. TheRMI was paid for one year and was renewable under the condition that the requisites of access to thescheme persisted.

STEFANO SACCHI AND FRANCESCA BASTAGLI 87

The monetary component

The RMI cash transfer is equal to the difference between the established threshold (adjusted with the ISEEequivalence scale) and the disposable income of a beneficiary household. In 2000, a single beneficiary withno income was entitled to an RMI cash transfer of � 268 per month, while the benefit for a family of fouramounted to the difference between � 660 and the disposable income of the family (see Table 3.14).

A uniform method for determining a household’s disposable income had been outlined by the decree thatlaunched the RMI but individual municipalities (responsible for the programme management andimplementation) generally made departures from the general rule in order to allow specific localcircumstances to be taken into account.

Disposable income was given by the total income perceived by an individual and his/her family includingcohabitants, adjusted for the following disregards and exceptions. Earnings from work were counted in totalhousehold income only for 75 per cent of their total amount. This reflects an attempt to avoid possibleunemployment and poverty traps. Some municipalities, moreover, decided to apply the same 25 per centdisregard to pension income. Although this was justified by considering pensions as deferred wages, it isdifficult to see a work incentive effect in the inclusion of pensions. Full disregard was applied to non-taxable social protection benefits, such as the family allowance, the benefit to large families and thematernity cheque. Many municipalities also chose to deduct a number of expenses in the calculation of apotential beneficiary’s income. The most common deductions include:

Figure 3.6 Municipalities involved in the first and second RMI experimentation phase.

88 ITALY: MINIMUM INSERTION INCOME

• rent about half of the municipalities decided to deduct rent. Some have done so completely (as in thecase of Naples and Limbiate) or by

Table 3.14 RMI threshold in 2000 according to household size (in euros)

Number of household components RMI threshold

1 268

2 420

3 546

4 660

5 764

For each additional component 94

fixed percentages (e.g. 50 per cent Alatri), others have deducted up to a fixed amount (e.g. � 155 in Catania,310 in Massa, 360 in Nichelino);

• mortgage rates on the house of residence, e.g. Cologno Monzese;• medical expenses not paid for by the National Health Service.

As far as the household’s wealth is concerned, any amount of assets (financial or real estate), with the exceptionof the house of residence, would have been a sufficient condition for losing eligibility, as determined by theinstitutive decree. Most municipalities, however, decided not to follow this regulation and disregardedmodest financial or real estate assets. Since owning the house of residence did not impinge upon eligibility,the decision on behalf of most municipalities to deduct rent from the household income is an equitable one,otherwise poorer families, which do not own their house, would be treated unfairly.

All these departures from the original law were designed to allow municipal authorities to respond todifferent social needs existing in different local realities. However, the evaluators report on the tendency forthe differentiated treatments to arise, in numerous municipalities, not so much in response to different needsbut rather for other reasons. The ratio between available resources and number of claims submitted for theRMI, for example, is one of the factors that determined whether a municipality chose to apply incomedisregard or deductions. The evaluators suggest that derogations were applied more frequently bymunicipalities in the North, while in other municipalities, typically in the South, where the number ofclaims was higher and the available resources are insufficient, fewer derogations were made.

The activation component

The receipt of the minimum income was conditional on joining a re-insertion or activation programme,devised by the local authorities, ideally designed around the personal characteristics of the beneficiary andaimed to promote individual capacities and financial independence. For underage beneficiaries theprogramme entailed compulsory education in the first instance and then vocational training. One remarkablefeature of the Italian RMI was that, at least in principle, every member of a beneficiary household wasrequired to participate in an activation programme. The refusal to commit to an insertion programme or thefailure to respect such commitment led to the suspension, reduction or revocation of the RMI.

The conditionality for beneficiaries of working age, non-employed and able to work was to take part intraining courses and to accept any job offer. This was ensured by the compulsory registration at JobPlacement Offices. Beneficiaries who were involved in education, vocational training, child care (of

STEFANO SACCHI AND FRANCESCA BASTAGLI 89

children younger than 3 or with a disability) or in rehabilitation programmes that do not allow workactivity, were not required to register at Job Placement Offices.

Other obligations beneficiaries were required to satisfy include informing the municipality of anymodification in family composition, income, assets and residence. Municipalities could suspend or reduce,also gradually or temporarily, the RMI according to the type of violation of the obligations listed. As forbeneficiaries who declared false information regarding their eligibility, in addition to suspension from theprogramme, they were required to return the monetary transfers received and could face penal charges.

The activation component of the RMI included a variety of services aimed at the development of theabilities of beneficiaries and at the reconstruction of social networks. The programmes offered by themunicipalities during the first phase of the experimentation were of the following kind:

• occupational (e.g. apprenticeship, internship, insertion in protected realities);• of public utility (e.g. maintenance of public parks, administrative work in municipal offices);• training (e.g. elementary training such as foreign language and computer skills courses, and vocational

training—courses to become a cook or an electrician, for instance);• educational (e.g. alphabetization, high school diploma);• rehabilitation (for beneficiaries with disabilities and problems of substance abuse);• caring and family support (e.g. education in personal hygiene, help in recovering from arrears, and

support for parental responsibilities, especially for single mothers);• social integration (e.g. social voluntary work, activities of social aggregation).

Financing the RMI

By law, 90 per cent of the RMI cash transfers were to be financed by the National Fund for SocialPolicies.67 Up to a maximum of the remaining 10 per cent was to be funded by the RMI municipalitiesthemselves. The RMI administration costs, including those of setting up and managing the integrationprogrammes, were to be financed entirely by the municipalities.

The central government had appropriated about � 246,000,000 for the first two years of RMIimplementation. This amount was allocated to the municipalities on the basis of estimates of potentialclaimants. Since in many cases the estimated and actual number of claimants differed, funds wereredistributed among municipalities.

At the end of the first two years of experimentation, total expenditure on the monetary component of themeasure amounted to � 220,270,000 approximately. By 31 December 2000, thirty out of the thirty-ninemunicipalities involved had not spent the entire monetary transfer obtained from the central government: atotal of � 26,490,000 had not been spent. These remaining funds were used to contribute to the financing ofthe second phase of the RMI. Overall, more than 97 per cent of the cash transfers paid to the RMI familieswas financed by the National Fund for Social Policies, but there is a remarkable variance among themunicipalities as regards their relative effort. On the one hand, fourteen municipalities did not contribute atall, on the other, ten municipalities shared the financial burden for more than 5 per cent of the expenditure,and two contributed with the maximum amount allowed, that is 10 per cent of the expenditure.

RMI beneficiaries

Between 1998 and 2000 the experimentation of the RMI involved nearly 35,000 households (seeTable 3.15). The vast majority of households benefiting from this scheme, more than 90 per cent of the

90 ITALY: MINIMUM INSERTION INCOME

total, resided in the southern municipalities. This figure should be assessed keeping in mind that 84 per centof the households living in a municipality involved in the experimentation resided in the twenty-foursouthern municipalities. On average, 4.2 per cent of the households residing in an RMI municipalitybenefited from the measure, but this figure conceals substantial regional variance: only 1.7 per cent ofhouseholds residing in one of the five northern RMI municipalities took advantage of the scheme, asopposed to 3.6 per cent in the Centre and 4.5 per cent in the South. This might be explained by the fact that,although all thirty-nine municipalities

Table 3.15 RMI claimant and beneficiary households (1998–2000)

Residentpopulation*

Residenthouseholds*

Claimssubmitted

Beneficiaryhouseholds**

Beneficiary/Residenthouseholds (%)

Accepted/Submittedclaims (%)

North 217,793 83,781 2,050 1,466 1.7 71.5

Centre 143, 112 49,195 2,674 1,789 3.6 66.9

South andislands

2,000,877 695,285 50,798 31,475 4.5 62.0

Total 2,361,782 828,261 55,522 34,730 4.2 62.6

Source: IRS, Fondazione Zancan, Cles (2001).Notes* 1997.** Beneficiary households=Claims accepted.

chosen were among those displaying higher unemployment and poverty, poverty rates in the South aremuch higher than in the North, and the group of potential RMI beneficiaries in the South is larger.However, some peculiar situations stand out: in Orta di Atella (Campania) nearly half of residenthouseholds were granted the RMI. Roughly the same can be observed in Isola Capo Rizzuto and Cutro(Calabria), where more than four households out of ten received the RMI. As for the ratio betweensubmitted and accepted claims, its average is about 63 per cent, with some variance both between andwithin regional groups of municipalities.

On 31 December 2000, total beneficiary households numbered over 25,000 and RMI beneficiariesamounted to nearly 86,000 (Table 3.16). Some 93 per cent of these lived in the South. The average numberof RMI family members was 3.35, with southern RMI households larger (nearly 3.5 members perhousehold) than northern ones (about 2.5).

By far, the prevalent family typology among beneficiaries was the couple with children: 60.5 per cent(Table 3.17). As a whole, households with children made up nearly 79 per cent of total beneficiaryhouseholds. However, given the predominance of southern families among the RMI beneficiaries, the totaldistribution largely reflects that pertaining to the former. In actual fact, the situation among householdsliving in the North (and, to a great extent, in the Centre) is fairly different. More than 43 per cent of northernbeneficiary households had no children (about 36 per

Table 3.16 RMI beneficiary households and individuals on 31 December 2000

Beneficiary families Beneficiary indivisuals Average number of members per beneficiaryhousehold

North 939 2,415 2.57

Centre 1,290 3,406 2.64

STEFANO SACCHI AND FRANCESCA BASTAGLI 91

Beneficiary families Beneficiary indivisuals Average number of members per beneficiaryhousehold

South and islands 23,362 79,997 3.42

Total 25,591 85,818 3.35

Source: IRS, Fondazione Zancan, Cles (2001).

Table 3.17 Composition of beneficiary households on 31 December 2000 (% values per row)

Single Couple withoutchildren

Couple withchildren

Couple withchildren and othermembers

Single parent Others Total

North 31.0 7.8 27.1 1.8 27.9 4.5 100

Centre 27.2 6.4 38.5 3.4 22.2 2.3 100

South and islands 12.4 5.3 62.8 3.7 13.7 2.1 100

Overall 13.6 5.4 60.5 3.7 14.6 2.2 100

Source: IRS, Fondazione Zancan, Cles (2001).

cent in the Centre) and the incidence of couples with children on total beneficiaries declines dramaticallythe more northwards one looks, while that of single parents and singles increases.

As for the employment status of RMI beneficiaries, the vast majority of beneficiaries were, almostequally, either unemployed or outside the labour force. Only 5 per cent of beneficiaries were employed, or atleast employed in the regular sectors of the economy (Table 3.18). This high coincidence between thecondition of beneficiary and that of being unemployed, on the one hand, seems to confirm data indicatingthe strong relation between unemployment and poverty, on the other, it raises some significant questionsconcerning the role of the RMI and its place within the broader welfare state framework. More specifically,the RMI was designed to act as a social safety net, a measure of last resort for households in a situation ofextreme financial difficulty and not as an instrument of income maintenance for the unemployed. Indeed, inthe scenario imagined by the Onofri Commission, the introduction of the RMI was envisioned as beingaccompanied by a reformed system of more effective and comprehensive social shock absorbers.

The insertion programmes

More than 37,000 beneficiaries (over 43 per cent of total beneficiaries) were involved in reinsertionprogrammes by 31 December 2000. The situation in the thirty-nine different municipalities variedsignificantly, with the percentage of individuals inserted in programmes varying from 8.4 per cent inCaserta, 10.5 per cent in Orta di Atella and 12.8 per cent in San Giovanni in Fiore, to 91.6 per cent in Massa,91.8 per cent in Nichelino, 93 per cent in Gallese and 100 per cent in Monterosi, Isernia, and Naples. If thisfirst RMI phase involved fourteen households in Monterosi and 202 in Isernia, in Naples it involved 3,695households and 17,336 individuals, all of which were registered as activated. These percentages raise somequestions and require further investigations on the homogeneity of the criteria adopted in administrating andimplementing the RMI. As for

92 ITALY: MINIMUM INSERTION INCOME

Table 3.18 RMI beneficiaries: status in relation to labour market on 31 December 2000

Status in relation to labour market % RMI beneficiaries

Employed 5.0

Unemployed 46.2

of which in search of first employment 10.5

Not part of workforce 48.8

Total 100

Source: Irs, Fondazione Zancan, Cles (2001).

Naples, for instance, the reported figure refers to the total number of members of those households in whichat least one member was assigned to a social worker and to a third-sector tutor, signing an insertion pact(PACI: Patto di aiuto concordato e individualizzato). Since a pact of this sort seems to have been signed inevery household, the total number of beneficiaries is recorded as activated. Given the quantitativeimportance of the figures relative to Naples (20 per cent of total beneficiaries), in what follows two kinds ofdata will be presented: the overall data, and data purged of Neapolitan beneficiaries. Excluding Naples fromdata dramatically brings down the total share of activated beneficiaries from more than 43 per cent to lessthan 29 per cent (Table 3.19).

As Table 3.19 shows, once again territorial differences are striking. Almost two-thirds of beneficiariesliving in northern municipalities and more than half of those living in the Centre were involved in aninsertion programme. Figures pertaining to the southern beneficiaries vary greatly according to whetherNaples is included or not. If it is, the share of activated over total beneficiaries is of 42 per cent approximately,still more than 20 percentage points lower than in the North and more than 15 than in the Centre; if it is not,only a quarter of southern beneficiaries were involved in an insertion programme after as much as two yearssince the beginning of the experimentation.

Table 3.20 shows the distribution of insertion programmes among programme types. The programmesmost frequently undertaken are those called “social integration”: 8,783 beneficiaries were involved in thistype of programme (about 24.5 per cent of total activated beneficiaries),68 The vast majority of these (6,905) resided in Naples, to the extent that if this municipality is removed, the share of beneficiaries involvedin social integration programmes is reduced to a mere 10 per cent. Social care and family support activitiesare also frequent: 7,351, making up more than 20 per cent of the total. Again, the high number ofNeapolitan beneficiaries involved in these programmes (5,050) lowers the figure to less than 15 per centwhen Naples is not included. However, this kind of programme can

Table 3.19 Beneficiaries involved in insertion programmes on 31 December 2000

Beneficiaries Activated beneficiaries Activated over total beneficiaries (%)

North 2,415 1,549 64.1

Centre 3,406 1,985 58.3

South and Islands 79,997 33,553 41.9

South and Islands without Naples 62,661 16,217 25.9

Total 85,818 37,087 43.2

Total without Naples 68,482 19,751 28.8

Source: Irs, Fondazione Zancan, Cles (2001) and authors’ own calculations on such data.

STEFANO SACCHI AND FRANCESCA BASTAGLI 93

Table 3.20 Distribution of activated beneficiaries among insertion programme types (% values per row)

Overall Without Naples

Occupational 14.9 15.9

Public utility 9.6 18.6

Training 11.6 15.4

Schooling 14.5 19.5

Rehabilitation 2.3 4.1

Caring and family support 20.5 12.4

Social integration 24.5 10.1

Other programmes 2.1 4.0

Total 100.0 100.0

Source: Irs, Fondazione Zancan, Cles (2001) and authors’ own calculations on such data.

be very effective in helping beneficiaries back to a normal life: nearly 1,500 instances concernbeneficiaries who were helped to plan the recover from a situation of indebtedness and eventually managedto pay arrears.

The relatively low incidence of occupational programmes (about 15 per cent) is striking. In manysouthern municipalities, they were not even present. When they were, their activation component wassometimes understood in a peculiar way (as occurred in Foggia, where beneficiaries pertaining to this typeof programme made up more than 60 per cent of the total, but were actually involved in training activities).

Programmes of public utility were mainly present in the South, but not in Naples, so that their figurealmost doubles (from 9.6 per cent to 18.6 per cent) when the latter municipality is removed. It seems hard toenvisage how such programmes might constitute a channel into the non-subsidized labour market.However, the difficulty in providing effective occupational programmes is not only linked to thecapabilities of the municipalities and their staff, but also to the labour market and the social networkconditions a municipality is embedded in. It is very likely that smaller municipalities could not cope with sucha task, relying on their own forces alone. This suggests that the municipal level might not always be themore appropriate administrative level at which to deal with such reinsertion efforts.

Two other types of programmes offered by municipalities were training and schooling. If the trainingprogrammes in many cases were exposed to the same problems as the occupational ones, schoolingprogrammes (which involved either 14.5 per cent or 19.5 per cent of beneficiary, depending on Naples isconsidered or not) seem to have been rather successful.

Programme administration and implementation

The RMI decree determined that the responsibilities of programme administration and implementationbelonged to the municipalities. This provision was understood in an extensive way by the municipalities,which considered it, for example, their right to allow some departures from the general rule as far as thecomputation of disposable income was concerned. In more general terms, the administration of the RMIwas embedded in the same logic underlying Law 328 of 2000. According to this Framework Law, thecentral government was responsible for the definition of the criteria of beneficiary identification and theessential and uniform levels of intervention, and for the allocation of financial resources of the NationalFund for Social Policies. The regions were responsible for the planning, coordination and guidance, whilethe monitoring of the territorial socio-economic situation was the responsibility of the provinces. Finally,

94 ITALY: MINIMUM INSERTION INCOME

according to the law, the municipalities played perhaps the most crucial role in the phase ofimplementation. It was their responsibility to identify strategic objectives and priorities and to developsupport networks, collaborations with third-sector associations and channels of communication.

The assessment of eligibility (identifying beneficiaries)

To apply for the RMI, a claimant had to sign a declaration indicating that he/she satisfied the requisites andconditions of eligibility. This procedure of “self-reporting” was introduced in 1997 as part of a sweepingreform aimed at simplifying many administrative procedures. On the one hand, this procedure has certainlyreduced time-consuming and costly processes for claimants, such as those involved in obtaining officialdocuments and information from various institutions. On the other, in the case of the RMI, self-reportingincreased the burden on administration since, following the submission of an application and beforeaccepting it, municipalities were responsible for testing the veracity of the information declared byclaimants.

The law establishes that those citizens who have declared false information may be prosecuted. However,while emphasizing the need for assessments and controls against fraud and abuse, the RMI legislation didnot vest municipalities with the authority to undertake investigations on income and assets property.Municipal authorities could only verify information held by the registry office such as whether anindividual resides in that municipality or the number of family members.

Most municipalities declared that they attempted in vain to mobilize those institutions that have thepower to undertake controls on income and assets. As a consequence, a widespread sense of impotenceamong administrative staff of the municipalities was reported, emerging from the impossibility ofundertaking effective examinations of the declarations of claimants. In some cases, however, the fiscalpolice (Guardia di Finanza) was mobilized. In the province of Enna, for example, the fiscal police openedan inquiry on suspicious cases and denounced 859 individuals for fraud and for declaring false informationin the presentation of personal documents.69

Administrating the applications

In advance of the launch of the RMI, Istat (the National Bureau of Statistics) provided estimates of thenumber of poor households in each municipality. In twenty-nine municipalities out of the thirty-nine thevalues were over-estimated: the number of actual submitted claims was lower than expected. In ColognoMonzese and Limbiate (Mi), the number of claims was over-estimated by more than 80 per cent. In othermunicipalities the discrepancy was smaller, as in the cases of Foggia, Barrafranca or Catania (about 10 percent). Low take-up due to eligible households not stepping forward to claim the benefit is one explanationfor these over-estimations. In ten municipalities the estimated number of potential claims was lower thanthe actual turnout. In these cases it is likely that some households claimed the RMI without satisfying theeligibility criteria, although this is only one reason for the under-estimation. Naples is one example, with anunder-estimation of 209.7 per cent. In this particular case, the under-estimation is to be attributed in part tothe design of the exercise: the estimates were made for certain areas of the city, those in which, initially, theRMI was to be introduced, while the programme was eventually extended to the whole city.

STEFANO SACCHI AND FRANCESCA BASTAGLI 95

Administrating the activation programmes

The municipalities were responsible for formulating, within thirty days from accepting the RMI claims, thepersonalized programmes of reinsertion. These programmes, devised by the municipal authorities togetherwith the single individuals concerned, in principle took into account the characteristics of all the householdmembers (every member of the beneficiary household had to agree to take part in an integration programme).

The most common type of collaboration between service providers was set up between municipalauthorities and voluntary associations in the elaboration and provision of “social integration” programmes(collaboration in eighteen municipalities), followed by the “rehabilitation” programmes provided by localauthorities with other public institutions. Networks were established between municipal authorities and otherpublic institutions in providing “schooling and education” programmes in sixteen municipalities. Thesepositive synergies are to be contrasted with the very low participation of the private sector (the EvaluationReport highlights the collaboration in four municipalities on occupational programmes and in threemunicipalities on training programmes).

The rate of participation of beneficiaries in the reinsertion programmes depended largely on the timing ofthe introduction or implementation of these programmes. The smaller the time-gap between the admission ofclaimants to the RMI and the actual launch of the insertion programmes, the greater the chance thatbeneficiaries were involved in the programmes.

Dealing with the underground economy

It is common knowledge that the underground economy plays an important role in Italy, particularly in theSouth. In order to address this issue in assessing eligibility, a number of municipalities devised someprocedures which certainly deserve to be analysed. When faced with a null or suspiciously low reportedincome, some municipalities automatically imputed a flat rate income to the claimant: Caserta, for instance,imputed � 2.58 per day for a single member household, plus � 1.29 per day for each additional householdcomponent. Other municipalities referred to the standard of living—assessed through phone bills, bankaccount statements, the possession of automobiles, etc.—in order to impute a presumptive income to theclaimant.70 Finally, in order to exclude non-deserving claimants from the benefit, some municipalities madeuse of the insertion programmes. They did so by requiring the beneficiary to participate in the programme atstrategic points of the working day, so as to make it impossible for him or her to attend another job in theunderground economy at the same time.

Exits from the programme

The evaluation report provided information on households and individuals that exited the RMI programme.Three main factors were associated with the interruption of the RMI cash transfer and integration programme:

• the situation of need was overcome (as a consequence of finding employment, for instance);• the beneficiary did not take part in the programme of insertion;• other reasons (such as change of residence, or death).

The evaluators reported that approximately 10 per cent of the households admitted to the RMI left themeasure within the two years of experimentation (see Table 3.21). However, there are discrepancies in thedata at this point. If, in the two years of experimentation, 34,730 claims were accepted —that is, 34,730households benefited from the RMI—and at the end of

96 ITALY: MINIMUM INSERTION INCOME

Table 3.21 RMI households and the reported exit from the RMI, reasons for ending benefit

Reported exits1999–2000

Reported exitsover totalacceptedclaims 1999–2000 (%)

Reported exits during experimentation, by reason (rowpercentages)

Condition ofneed overcome

RMIabandoned

Other motive Total

North 469 32.0 53.7 24.1 22.2 100

Centre 506 28.3 56.3 26.5 17.2 100

South 2,199 7.0 53.7 9.0 37.3 100

Total 3,174 9.1 54.1 14.0 31.9 100

Source: Authors’ own calculations based on Irs, Fondazione Zancan, Cles (2001) data.

the experimentation the number of households still benefiting from the measure amounted to 25,591, it shouldbe the case that 9,139 households entered the programme and then exited from it, that is about a quarter (26.3 per cent) of total accepted claims, and not a mere tenth (3,174 households) as reported. It is possible thatthe latter figure (3,174) is made up only of the number of exits from the programme the municipalitiesreported, and not of the total exits.71 Be that as it may, this should simply warn the reader against drawingrushed inferences from the data, as it should have done with journalists and policy-makers commenting on asupposed RMI failure.

Among the reported exits from the RMI, the main reason is always the overcoming of the condition ofneed, which applies to more that 50 per cent of the exits. As for the abandonment of the RMI as aconsequence of not taking part in the insertion programme, the incidence of this reason of exit seems to bemuch higher in the northern and central municipalities than in the South (conditional on the available data).This could simply be connected to the non-availability of insertion programmes (if programmes are not putin place, beneficiaries cannot be sanctioned), rather than to the preference of southern local authorities forsofter negative incentives towards beneficiaries who do not comply. However, the available data do notallow any conclusions to be drawn.

All these data, however sketchy, refer to the insertion side of the RMI. Nevertheless, the RMI is not to beconfused with a labour market policy. Its prime objective is to provide income support to poor householdsand help develop processes of social integration. As such, its effectiveness should be assessed, at least in thefirst instance, in terms of enhancement of the beneficiaries’ daily life.

Unfortunately, no data are available to assess the RMI qua poverty alleviation policy. This is aninformation gap that should be addressed by future social policy initiatives more broadly. One lesson thatemerges from the RMI experience is the pressing need to improve data collection systems so as to ensureadequate data necessary for any sound monitoring and evaluation activities.

Conclusion

As the preceding sections have shown, the past decade has witnessed the development of a social inclusionstrategy, which has then ground to a halt. As the main building block of such strategy, the Reddito minimo diinserimento and its vicissitudes epitomize this trajectory. However, just as the disappearance of itsinstitutional and symbolic cornerstone has made the fight against poverty and social exclusion in Italy wither,

STEFANO SACCHI AND FRANCESCA BASTAGLI 97

the introduction of some sort of minimum income programme could by the very same token infuse it withnew energy.

Although the constitutional reform of 2001 has taken away from the state considerable steering power inthe social assistance field, still it has reaffirmed its competence in setting the essential levels of provision.This is where a minimum income scheme inherently belongs, at least in its monetary, income guaranteecomponent.72

As we saw, the Budget Law for 2004 provided for some sort of financial contribution on the part of thestate to those regions which introduce the Income of Last Resort, RUI. Although at the time of writing thecontours of this measure are still surrounded by thick mist, it does not seem to be shaping in the hoped-fordirection. Despite its ambiguous phrasing, the Budget Law would seem to rule out the possibility that theRUI is listed among the essential levels of provision in social assistance.73 If this is the case, though, itfollows that individual regional governments are unconstrained in their decision whether to introduce theRUI or not, since the state cannot encroach upon their constitutional rights by imposing the RUI on them. Ifa regional government should introduce the RUI, then the state would have to help funding it, given theprovisions in the Budget Law, but each regional government would seem to have the constitutional right notto establish it. The danger that the RUI may be established only in those regions that have theorganizational capability and the financial means to run it is all too evident. Given the geography of povertyin Italy and the degree of both autonomy and capacity of public authorities in some southern regions, thiswould entail dire consequences.74

However, that a safety net is not included in the list of essential levels of provision for the time being,does not rule out the possibility that this could be the case in the future. If this were to happen, thedifficulties that emerged during the experimentation of the RMI would have to be addressed, lest theyshould hamper the effectiveness of the new measure.

The critical factors that plagued the RMI experimentation can be classified into three groups: (1) thedesign of the monetary component; (2) the lack of administrative capacity and resources on the part of manylocal authorities, adversely affecting the design and implementation of inser tion programmes and theadministration of means testing; and (3) the risk of functional overload of the minimum income scheme.

The main flaws in the design of the monetary component were the lack of consideration of territorialvariance in the cost of living, the erratic treatment of wealth, variance and even logical mistakes in theapplication of disregards (such as the 25 per cent disallowance on pension income).75 In many localcontexts the derogations made to the national guidelines, although somewhat justified by the necessity ofaccommodating local circumstances within such rigid national framework, led to the inclusion of aconsiderable share of the population in the pool of beneficiaries, thus jeopardizing the reinsertion aim of themeasure with plausible work disincentive effects.

All these flaws of the monetary component of the RMI can and must be fixed by aptly designing themonetary component of any prospective minimum income scheme. Some arrangements, such as theinclusion of wealth in the means testing through the adoption of the ISEE, or the consideration of thevariable cost of housing according to the geographical residence of the beneficiary, or better labour supplyincentives, can be relatively easy to craft (Sestito and Nigro 2004). However, the over-inclusion ofbeneficiaries in some contexts points to the two other challenges that emerge from the RMI experience, andwhich are far more difficult to resolve with institutional design alone: the low degree of institutionalcapabilities and resources of many local authorities, and the risk of using the safety net to cater for too manyunmet needs.

The experimentation of the RMI allowed social workers to gain awareness of situations of deprivationand exclusion not formerly known and provided them with a policy tool to deal with such situations. In a

98 ITALY: MINIMUM INSERTION INCOME

number of municipalities, the introduction of the RMI fostered the reorganization of the existing system oflocal welfare policies towards its rationalization and integration. At the same time, though, manymunicipalities lacked both capability and staff required in order to administer the programme, both in termsof designing and running insertion projects and in terms of administering means testing.

As Maurizio Ferrera writes in his Introduction to this volume, means-tested schemes are very demandingin terms of institutional capabilities and managerial skills. Given the peculiar characteristics of thesocioeconomic environment that generates demand for means-tested benefits in the (southern) Italiancontext, the managerial skills to be developed are particularly sophisticated. True, valuable help can come tolocal administrators by standardized means testing instruments and arrangements, such as the ISEE and theautomatic imputation of a flat-rate presumptive income. By the same token, more collaboration on the partof fiscal administration and police, and provisions that should vest the local authorities with the power toundertake controls on income and assets property may be very useful indeed. However, given the externalpressures many administrators find themselves subject to in the South (the lack of autonomy) and thewidespread use of social benefits for patronage exchanges between politicians and voters (Ferrera 1996), itis fair to admit that this is not at all an easy problem to overcome, and all the more so if weaker regions andlocal authorities are not supported by the central level.

Evidently, administrative capabilities are a critical factor for the insertion programmes also. The RMIexperimentation has shown that it might prove impossible for smaller municipalities to devise and manageuseful programmes, due to lack of staff and competence, but also it might reveal as beyond their reach theactivation of those networks among the socioeconomic actors necessary to bring about positive results. Thisraises the question of whether the municipal level is always the best one in administering the measure, andif it would not be better to allocate the responsibility for the day-to-day management of the measure tosufficiently large territorial entities such as those zone foreseen by the Framework Law 328 of 2000.

There is another aspect highlighted by the RMI experimentation as regards insertion programmes that isworth looking at, though. A cursory glance at the score of activated over total beneficiaries in the North andthe Centre, on the one hand, and the South on the other (Table 3.19) is sufficient to bring to the fore theissue of the functioning of a safety net given the socio-economic environment within which poverty isembedded in Italy’s Mezzogiorno. This is relevant for insertion programmes: how can local administratorsmanage to set up social insertion or activation programmes in an economically depressed environment? Butit is even more relevant for the safety net in its entirety: without a functioning labour market, a minimumincome scheme risks functional overload. From a last resort option, catering to the specific needs ofmarginal groups, it tends to become “the only game in town”, as Ferrera aptly notes in his introductory chapter,supposedly expected to solve all problems.

It may well be the case, thus, that when faced with all the problems and risks of crafting andimplementing a minimum income programme in the Italian context, instead of including such a safety netwithin the essential levels of provision concerning social rights Italy’s policy-makers could refrain from theenterprise and opt once again for the well-known, less demanding categorical measures. In our view, thiswould be a mistake.

We are perfectly aware of the risk of institutional slacks and functional overload connected to a minimumincome scheme in Italy, even after a good part of it has been neutralized by careful institutionalengineering. But considering the current poverty situation in Italy, the utter ineffectiveness of existingpolicies in reducing poverty in the face of the huge amount of monies poured into them (think of invaliditypensions), their appalling iniquity and their exploitation for clientelist exchanges and patronage purposes(think again of invalidity pensions), we maintain it is a risk well worth taking.

STEFANO SACCHI AND FRANCESCA BASTAGLI 99

Appendix

Table 3.22 Chronology of legislative provisions in the field of poverty alleviation

Legislative Act Content

1952

Law 4 April 1952, no. 218 Introduction of the pensione minima and related pensionssupplements

1969

Law 30 April 1969, no. 153 Introduction of the social pension, a social assistancemeasure for over-65 year-olds with an income below aset threshold

1977

DPR 24 April 1977, nos 616 and 617 Transfer of legislative, planning and organizationalresponsibilities regarding social assistance measures tothe regions and of administrative and managementfunctions to municipalities

1984

Law 22 June 1984, no. 222 Revision of the discipline of the insurance-based (INPS)invalidity pension

1988

Law 13 May 1988, no. 153 Revision of insurance-based family allowances;henceforth conditional on a family income threshold, seton the basis of number of family members and on thepresence of particular difficult circumstances within thehousehold

1997

Law 15 March 1997, no. 59 and Law 15 May 1997, no.127 (Bassanini Reform)

Transfer of competences and responsibilities to regionsand local authorities; reform and simplification of publicadministration; Simplification of administrativerequirements

Law 27 December 1997, no. 449 (Budget Law for 1998) Introduction of the National Fund for Social Policies;Provision of the Minimum Insertion Income (RMI)

1998

Legislative Decree31 March 1998, no. 109Legislative Decree18 June 1998, no. 237

Introduction of the Indicator of the Equivalent EconomicSituation (ISEE) Introduction, in an experimental fashionand in some areas, of the minimum insertion income, RMI(Reddito Minimo di Inserimento)

Law 23 December 1998, no. 448 (Budget Law for 1999) Introduction of the assistance-based benefit for familieswith at least three children and of the maternity cheque,both means-tested through the ISEE

Legislative Act Content

2000

Legislative decree 3 May 2000, no. 130 Law 8 November2000, no. 328

Final adoption of the ISEE with definite criteria as regardshousehold composition Framework Law for theRealization of an Integrated System of Social

100 ITALY: MINIMUM INSERTION INCOME

Legislative Act Content

Interventions and Services (“framework law” or “law onthe reform of social assistance” in the national debate)

Law 23 December 2000, no. 388 (Budget Law for 2001) Extension of RMI experimentation

2001

Presidential decree of 3 May 2001 Issue of the National Plan of Social Interventions andServices as established by Law 328/2000

Constitutional law 18 October 2001, no. 3 Law 28December 2001, no. 448 (Budget Law for 2002)

Reform of Title V of the Constitution (Titolo V), whichestablishes the legislative powers of the various levels ofgovernment Increased maggiorazioni sociali for theover-70s; increased tax credits for dependent children

2002

Law 27 December 2002, no. 289 (Budget Law for 2003) Introduction of the no tax area

2003

Law 24 November 2003, no. 326 Benefit for each second child

Law 24 December 2003, no. 350 (Budget Law for 2004) Provision that the state would contribute to financing theregions which establish the Income of Last Resort

Acknowledgements

This chapter has been jointly discussed and devised by the authors. The introduction, the second, third andfourth sections and the conclusions were written by S.Sacchi; the fifth section on the Reddito minimod’inserimento and the Appendix were written by F.Bastagli. While working on this chapter, FrancescaBastagli was at the Centre for Comparative Political Research POLEIS, at Bocconi University. The findings,interpretations and conclusions of this chapter are those of the authors and do not in any way reflect theviews of Bocconi University or the World Bank. We are grateful to Maurizio Ferrera for continuous supportand advice and to Tiziana Alti and David Benassi for valuable comments. We would also like to thank allthe participants of the FIPOSC seminar (Milan, 23 and 24 May 2002), where the first version of this chapterwas presented. The usual disclaimer applies.

Notes

1 Operationally, social assistance normally denotes those programmes aimed at guaranteeing a modicum ofresources to those who are in a manifest situation of need, and are subject to some form of means testing, that isof assessment of the impossibility of the claimant to cope with the situation of need with own economicresources (typically income, but sometimes also wealth). Often social assistance measures contain categoricalelements, that is they make the right to enjoy a given benefit conditional upon the beneficiary’s belonging to aspecified category (say, that of mothers with under-age children) identified by criteria different from merelyeconomic ones (see Toso 2000a). Sometimes social assistance denotes all means-tested measures with verticaland horizontal redistributive aims, irrespective of their being conditional on contribution requirements or not.What really counts, in this broad meaning of social assistance, is that social protection benefits are means tested.In what follows, we will not adopt this broad meaning. Social assistance will thus denote only those interventionsunconditional on contribution history. We argue that there is a conspicuous difference between benefits inprinciple accessible to every citizen (or even resident) and those of a social insurance nature, accessible only tothose who have worked for a long enough period of time in the formal labour market (and therefore reach core

STEFANO SACCHI AND FRANCESCA BASTAGLI 101

workers but exclude students, young people seeking first-time employment, young mothers, women withinterrupted careers, the elderly with insufficient contribution seniority, atypical workers, immigrants). Obviouslyenough, since the focus is here on poverty alleviation policies in Italy, we will also consider social insurancemeasures that have an impact on poverty, but their difference from social assistance policies will be highlighted.

2 With Law no. 6972 of 1890, known as legge Crispi.3 Since 1995, for new claimants this scheme has been replaced by a very similar one, the assegno sociale (social

allowance).4 Source: Ministero dell’Economia e delle Finanze (2003).5 The Appendix provides a chronological list of the most important legislative provisions in the field of poverty

alleviation, with a special focus on the most recent ones.6 On the concept of re-calibration, see Ferrera et al. (2000).7 As Boeri and Perotti (2002) rightiy point out, this was not an experiment at all, because there was no control

group, identical to (or randomly chosen as) the treatment group. Nor was it a demonstration, since it is true thatthe latter may or may not use controls, but still presupposes a research plan (see Focus, Institute for Research onPoverty, University of Wisconsin-Madison, vol. 5, no. 3, summer 1982). However, from its very beginning the“experimentation” phase was not intended to estimate the effect of the minimum income programme onbeneficiaries in order to decide whether to extend it to the whole Italian territory as a fully-fledged policymeasure. The effectiveness of the measure was taken for granted, and overall the introduction of a minimumincome scheme was considered necessary regardless of any experimentation, also for political reasons (amongthe fifteen EU member states of the time, Italy was—and still is —in the sole company of Greece in lacking sucha scheme). Thus, the “experimentation” was mainly set up for administrative and organizational purposes: was tohelp to identify the operational problems and difficulties met by local authorities in providing the benefit,managing insertion programmes, combating frauds, and so forth. In this respect, the RMI “experimentation” hasprovided policy-makers with a great deal of evidence that—albeit anecdotal— could still be rather helpful.

8 These measures are described on p. 104.9 The constitutional reform of 2001 (see below) has taken away from the central level of government the power to

issue planning instruments in the social assistance field, therefore the Social Plan that expired in 2003 could notbe replaced.

10 The National Fund for Social Policies was first introduced by the Budget Law for 1998 (Law 449/97) with theobjective of financing at a central and local level social policies aimed at children, the elderly, the disabled, theprevention of drug abuse and the rehabilitation of addicts, family support and the integration of foreign citizens.Law 328/2000 established that starting from the year 2002, the total appropriation of the Fund is determined bythe Budget Law and no longer by the sum of appropriations made by single laws. This should facilitate a morecoordinated programming of social policies at all institutional levels. The central government is also responsiblefor the allocation of the financial resources, which takes place after an agreement is reached within theConferenza stato-regioni, città e autonomie locali (the institutional venue comprised of all the various territoriallevels of government in Italy: state, regions, municipalities and local authorities). The total appropriation for theFund amounted in 2003 to � 1,716,500,000 approximately, 0.13 per cent of the 2003 GDP.

11 The strategy rests on the application of the Open Method of Coordination to the fight against poverty and socialexclusion and consists of two parts. One is a multi-annual action programme designed to encourage cooperationbetween Member States to combat social exclusion. The other is a process involving the submission, every twoyears, of National Action Plans on the part of member states and their assessment by the Commission and theSocial Protection Committee, which brings about a Joint Report of the Council and the Commission. The five-year action programme started on 1 January 2002 and will last until 31 December 2006, with a budget of � 75,000,000 for the whole period. The first round of the social inclusion process started in June 2001, when the first Naps/incl, setting out national strategies for the period 2001–03, were presented. The first joint inclusion report wasthen issued in December 2001. The second round started with Naps/incl 2003–05 in July 2003, leading to thepresentation of the second joint inclusion report to the Spring European Council in March 2004. The whole

102 ITALY: MINIMUM INSERTION INCOME

process is guided by the objectives in the fight against poverty and social exclusion approved by the EuropeanCouncil in Nice, in December 2000, that are:

1 To facilitate participation in employment and access by all to the resources, rights, goods andservices.

2 To prevent the risk of exclusion.3 To help the most vulnerable.4 To mobilize all the relevant bodies.

For a more detailed account of the social inclusion process, see Ferrera et al. (2002).12 Although presented in July 2001, after the general elections had given a large majority to the centre-right

coalition led by Silvio Berlusconi, Italy’s first Nap/incl is to be politically ascribed to the later centre-leftgovernment, led by Giuliano Amato, in office until May 2001. On Italy’s participation to the social inclusionprocess, see Ferrera and Sacchi (2004) and Sacchi (2005).

13 For a thorough account of such developments, see Ferrera and Sacchi (2004).14 The reform came into force after the confirmatory referendum of October 2001. However, it had been promoted

and voted by MPs under the centre-left governments.15 However, as argued by Gori (2003), it may well be a difficult task to set the essential levels of provision of non-

monetary benefits (and those regarding the prevention of exclusion and promotion of inclusion in particular).16 It is to be noted that it was the same centre-left parliamentary majority to endorse both the Framework Law and

the constitutional reform.17 The White Paper on Welfare (Ministero del Lavoro e delle Politiche Sociali 2003a) identifies two “new

priorities”: “to manage demographic change” and “to put the family at the core of policy action”. However, the waythe White Paper deals with the role of the family within the welfare system seems completely at odds with thenormative conclusions of contemporary social science research (see, for instance, Esping-Andersen 1999). Butthe most relevant feature of the White Paper on Welfare is that only general priorities are set, while the issue offinancial resources necessary to implement such (vague) priorities is avoided completely. For a discussion of theWhite Paper on Welfare, see the contributions by Del Boca (6 February 2003), Saraceno, Boeri and Perotti, andRanci (20 February 2003) on the web journal La Voce (www.lavoce.info, accessed 24 May 2004).

18 These interventions will be considered on p. 102.19 See Ferrera and Sacchi (2004) and Sacchi (2005).20 See p. 109.21 A possible objection has to do with Trattamento di fine rapporto (TFR) being entirely computed in the “old age

and survivors” function by EUROSTAT. More an accrued deferred wage than a form of severance pay, for eachdependent worker the TFR is calculated as a share of the worker’s annual wage, retained by the employer as asource of low-cost financing and then paid (with inflation compensation and interest) to the employee wheneverthe work relation ends. TFR is thus paid to dependent workers who retire, to those who quit their job, and tothose who lose their job. Only in the first case can TFR be considered analogous to old age pensions; in the lastcase, TFR can be considered to address the risk of unemployment, whereas in the case of workers who quit theirjob while at the same time entering a new work relation, TFR should not be included among social protectionbenefits, not being intended to relieve households and individuals of the burden of any defined risk or need.EUROSTAT (2004) estimates total expenditure on TFR as 6 per cent of total social benefits, thus in no waywould a different allocation of TFR even blur the overall picture as regards the skewness towards old age ofItaly’s social expenditure.

22 See note 1. Moreover, almost 46 per cent of total public expenditure on both cash and in-kind social assistancebenefits was absorbed by a single measure: civil invalidity pensions.

STEFANO SACCHI AND FRANCESCA BASTAGLI 103

23 With the exception of the very recent benefit for each second child, the only universal cash benefit in the Italianwelfare state is the care allowance (see p. 102). Outside the realm of cash benefits, the picture is different: forinstance, health care is universal in coverage.

24 A possible objection here is that Italy might well devote the bulk of its social expenditure to cover the risks of oldage (as opposed, say, to cater for the needs of families and children) simply because it has a particularly highshare of aged population. Indeed, Italy has a particularly high share of aged population: in 2000, the ratio of thepopulation aged 65 and over to the population aged 15–64 was as high as 27 per cent, compared to an EUaverage of 24 per cent (Economic Policy Committee 2001). But the value for Sweden was the same as for Italy,with an internal structure of social expenditure much less skewed towards old age in the former country than inthe latter: 39.1 per cent of total social benefits were allocated to the “old age and survivors” function in 2000 inSweden (as opposed to 63.4 per cent in Italy), while 10.8 per cent of total social benefits appertained to the“family and children” function (as opposed to 3.8 per cent in Italy). Therefore, allocation of social expenditureacross functions is a matter of policy choice (and policy legacy) in the face of environmental challenges.

25 Moreover, not only is expenditure on pensions disproportionate, but it has perverse redistributive effects too,benefiting high-income households. Boeri and Perotti (2002) show that the ratio between the share of pensiontransfers in disposable income for households with below-average disposable income and the same share forthose with above-average disposable income is the lowest across ECHP countries.

26 An absolute poverty line is usually taken to be one which is fixed over time in terms of purchasing power,allowing the consumption of a specified basket of goods and services representing for example the possibility ofsatisfying “basic needs”. A relative poverty line varies with changes in median or mean national income orexpenditure, a line fixed as a proportion of the median national income, for example.

27 A break in the series makes pre-1997 data non-comparable with those for 1997 and subsequent years.28 As had already happened between 2000 and 2001, between 2001 and 2002 the average living standard (assessed

through the average real consumption expenditure) fell, due to a fall in real consumption expenditure amongmore affluent households. A fall—in real terms—of the relative poverty line ensued, thereby causing an overallreduction in the number of households below the relative poverty line. This can be seen by comparing relativepoverty data for 2002 to figures for the same year derived from adjusting the 2001 relative poverty line for theestimated inflation rate, so as to keep the relative poverty line constant in real terms between 2001 and 2002. Theinflation-adjusted (constant in real terms) relative poverty line for 2002 amounts to � 844, roughly � 21 higher thanthe one calculated sticking to the International Standard of Poverty Line. As a consequence, applying the adjustedpoverty line the overall number of households plagued by relative poverty in 2002 would be roughly similar tothe figure for 2001, and so would the headcount ratio (11.9 in 2002 as compared to 12.0 in 2001). However, byno means does this constitute a poser for the following analysis, insofar as the focus of this section lies in singlingout the categories of households most affected by poverty—Italy’s poverty profile, so to speak, which seems toremain the same year after year.

29 Poverty data based on income may be obtained from the Bank of Italy sample survey of household income andwealth, undertaken every two years (see Brandolini 1999).

30 In what follows, the poverty headcount ratio will also be called incidence of poverty and poverty rate. Moreover,the average poverty gap ratio among the poor will be taken as a measure of poverty intensity, even though we areaware that the latter term has been used in recent technical literature to denote the Sen-Shorrocks-Thon index.

31 1.027 (female head) as opposed to 0.990 (male head), so the ratio between the two is practically 1 (1.036).However, this seems to disguise considerable territorial variance: the ratio between the relative risks of poverty(female/male) is 1.077 in the South, 0.774 in the Centre and 1.4 in the North, meaning that in the North ahousehold with a female head is significantly more at risk of being poor than one with a male head, while in theCentre the reverse obtains. Here and elsewhere relative risks of poverty are defined as the ratio between theincidence of poverty for a given category and that for the population at large. The ratio between two relative risksof poverty then equals the ratio between the poverty rates for the two categories taken into account.

104 ITALY: MINIMUM INSERTION INCOME

32 Data not reported here show that in 2000 individuals under 18 and over 65 were sharing the same poverty rate:16.7 per cent, compared to an overall individual poverty rate of 13.9 per cent. Source. Commissione d’Indaginesull’Esclusione Sociale (2002).

33 Data from the European Community Household Panel (ECHP). They were collected in 2001 and reflectrespondents’ income situation in 2000. ECHP limitations are well known—see Council of the European Union(2003) for a succinct account of ECHP methodology and limitations. Still, ECHP data are the only comparabledata across the EU at present day, albeit old and of questionable quality. The reference variable is equivalizeddisposable income. The poverty line is set at 60 per cent of national median equivalized disposable income. Theequivalence scale is the modified OECD scale, which counts the first adult (i.e. person aged 14 or over) as 1.0,each other adult as 0.5 and each child aged under 14 as 0.3. At persistent risk of poverty are those individualswhose equivalized disposable income falls below the poverty line in the reference year and in at least two of thepreceding three years.

34 Boeri and Perotti use data from waves I to III of ECHP, collected in 1994 to 1996 and referring to respondents’income from 1993 to 1995. Poverty threshold is set at 50 per cent of median equivalized income. Persistence inpoverty is defined here as being in poverty for three years in a row.

35 Table 2.3 in Boeri and Perotti (2002:56–57). Data come from wave IV of ECHP, reflecting respondents’ incomein 1996. As for the relative risk of poverty for households headed by an over-65-year-old as such, findings tendgreatly to vary depending on the choice of the reference variable: income or consumption expenditure, due toreduced consumption among the elderly. A study made by the Italian Commission on Social Exclusion applyingthe same procedural standard (ISPL) and equivalence scale (Carbonaro) to same-year (1998) household income(Bank of Italy Survey) and consumption (ISTAT Survey) data show that the relative risk of poverty forhouseholds headed by an over-65-year-old equals 0.766 when poverty is defined in terms of income (the lowestvalue across households by age of head) and 1.338 when poverty is defined in terms of consumption (the highestvalue across households by age of head). See Table 6.1 in Commissione d’Indagine sull’Esclusione Sociale(2002:124).

36 In fact, since 1 December 2003 (and, at least for the time being, until 31 December 2004) families can claim aflat-rate, one-off cash benefit of � 1,000 for any newborn or adopted child after their first one. Although thepurpose of this measure is to raise the number of births, it could nonetheless have poverty alleviation effects: seebelow.

37 Since 1998 this measure has been extended to some categories of the self-employed.38 For unmarried pensioners the income ceiling equals twice the yearly statutory minimum. As for married

pensioners, they must pass both the personal income test and the couple’s income test (the cumulated income ofthe couple must be lower than four times the yearly statutory minimum).

39 If the beneficiary is married, personal income must be lower than � 6,836 and the couple’s cumulated incomemust be lower than � 11,500 approximately.

40 Beneficiaries with modest incomes receive a benefit topped up to the social allowance.41 The pension amount is calculated adding to contribution seniority a “contribution bonus” that allows the

beneficiary to reach the pensionable age. If necessary, such amount is topped up to the minimum. Themaggiorazioni sociali normally reserved for over-70s are in the case of pensione di inabilità granted to thoseover 60.

42 Also beneficiaries of the pensione di inabilità civile can claim at 60 the higher benefits normally reserved forover-70s.

43 If the claimant is married, both income tests must be passed, that is the claimant’s personal income must be lowerthan � 3,846 even when the total income of the couple is lower than the � 13,253 ceiling.

44 Unlike the social pension, the social allowance is granted also when the married claimant’s income is higher thanthe personal income ceiling, provided that the total income of the couple is lower than � 9,334.

45 Approximately 80 per cent of recipient households resided in six regions: Campania, Sicilia, Puglia, Calabria,Basilicata and Sardegna.

STEFANO SACCHI AND FRANCESCA BASTAGLI 105

46 Although there are strong theoretical reasons for taking into account wealth in means-testing formulas (seeBaldini et al. 2000), this has been done for ISEE mainly for practical reasons, in order to avoid social assistancebenefits being enjoyed by income tax dodgers.

47 The ISEE Equivalence Scale is obtained by raising the number of household components to the power 0.65. Sucha basic scale is then increased when the following conditions apply: single parent households with under-agechildren (+0.2 points); households with under-age children in which both parents work (+0.2 points); householdcomponent with permanent handicap or seriously disabled (+0.5 points for each such component).

48 Baldini et al. (2000) carried out a microsimulation exercise concerning the distributive consequences of adoptingthe ISEE instead of taxable income alone as the reference variable for means testing. The introduction of theISEE would imply that nearly 13 per cent of those households eligible in terms of income would no longer beclassified as such in terms of the ISEE, while an identical percentage of households who are not eligible in termsof income would be entitied to some benefits according to the ISEE. To summarize, the gainers from theadoption of the ISEE

are likely to be households in the first stages of their life cycle and with a dependent worker as the head.On the contrary, the groups who run the greater risk of being excluded from social expenditure are likelyto be households with a head not employed as dependent worker (particularly the self-employed) or retired,with a high level of education, residing in their own house and with a high propensity to evade thepersonal income tax.

(ibid.: 15)

49 Data corrected in order to take non-reporting and misreporting into account.50 Poverty is defined with reference to equivalent disposable income before social transfers and the poverty line is

set at 60 per cent of the median of such income distribution. The equivalence scale is given by the number of familymembers raised to the power 0.65 (the coefficient used in the ISEE scale).

51 The outlays necessary to extend the Reddito minimo di inserimento to the whole national territory can be roughlyestimated as � 3,000 million per year, about 0.23 per cent of 2003 GDP (see also Sestito and Nigro 2004). In 2002expenditure on civil invalidity pensions amounted to � 9,500 million, 0.75 per cent of 2002 GDP.

52 Italy’s Nap/incl 2003–05 reports that “according to the first available evidence this measure has helped over 300,000 households to overcome the poverty threshold” (Ministero del Lavoro e delle Politiche Sociali 2003b:15),but it does not cite any verifiable source.

53 The no tax area is an above-the-line tax deduction, shielding a variable amount of earned income from theapplication of personal income tax. It consists of an across-the-board deduction of � 3,000 per year plus a taxdeduction which depends on the taxpayer’s professional situation and on the share of year he or she actuallyspent in work (or retirement). For employees and pensioners, the latter tax deduction is proportional to the shareof the year actually spent in work or in retirement, respectively, so that the maximum overall deduction amountsto � 7,500 per year for employees and � 7,000 per year for retired people. For the self-employed, the overall taxdeduction amounts to � 4,500, regardless of the share of year spent in work.

54 On the fact that RMI was not a proper experiment see note 7.55 They were: Milan’s Istituto per la Ricerca Sociale (IRS), Padua’s Centro Studi e Formazione Sociale Fondazione

Emanuela Zancan and Rome’s Centro di Ricerche e Studi sui Problemi del Lavoro, dell’Economia e delloSviluppo (CLES).

56 With a total appropriation of � 402,840,000 over the two years (� 220,000,000 for the year 2002 alone).57 Patto per l’Italia was signed by CISL and UIL, but not by CGIL. See Ferrera and Gualmini (2004).58 “Patto per l’Italia. Contratto per il lavoro. Intesa per la competitività e l’inclusione sociale”, point 2.7. The

document is available for download from the website of the Welfare Ministry (www.welfare.gov.it, accessed 24May 2004).

59 In short, involved municipalities were given an extended deadline (till the end of 2004) to allocate those fundswhich they should have exhausted by 2002, but that in many cases did not reach them until 2003. The extra

106 ITALY: MINIMUM INSERTION INCOME

funding to some municipalities (those that had already exhausted the funds) was decided primarily on thegrounds of public order considerations.

60 Art. 3, no. 101, Law 350 of 2004.61 See also Sestito and Nigro (2004).62 Legislative Decree 237 of 1998 specified that in selecting the municipalities the following factors must be taken

into account: (1) poverty levels; (2) variance of economic, demographic and social conditions; (3) variety ofexisting social assistance measures in each municipality; (4) adequate territorial distribution of the municipalitiestaking part in the experimentation; and (5) willingness of the municipality to participate.

63 During the 1990s new policies for local development were created in Italy: the so called territorial pacts. Suchpolicy measures are aimed at the creation of a specific local policy network, built by social partners, localinstitutions and associations. The policy network is in charge of the definition of a set of development goals (inparticular, the creation of new jobs through the funding of innovative business initiatives) which are implementedwithin the frame of a Patto territoriale by one or more of the local public institutions (usually municipalities orprovinces). The “bottom-up” approach underlying these policies is particularly innovative with respect to theformer Italian local development policies.

64 Electoral considerations, related to the overlapping of some territorial pacts with important constituencies, alsoseem to have been at work.

65 A more detailed description of the RMI experience can be found in the FIPOSC final report (FIPOSC 2002).Extremely useful are also the essays in Calza Bini, Nicolaus and Turcio (2003) and the special issues ofL’assistenza sociale, July–December 2002, nos 3–4 and of Prospettive sociali e sanitarie, July–September 2002,nos 13–15.

66 See note 47.67 Legislative Decree 238 of 1998.68 Since Catania did not provide the evaluators with even the slightest information about the types of programme

undertaken, the denominator here amounts to 35,887, the number of beneficiaries activated excluding those ofCatania.

69 However, the Welfare Ministry has not so far sued those individuals for damages. 70 In this regard, in some municipalities the municipal police visited the claimants’ homes in order to assess their

standard of living.71 Moreover, Catania did not provide the evaluators with any data about exits from the programme.72 To be sure, to craft the insertion component as an essential level of provision is an arduous task indeed (Gori

2003), even intractable when it comes to the exercise of substitutive power on the part of the state in order toensure that beneficiaries enjoy the insertion component. Still, the central level of government could devise aframework as regards such component, that the local authorities must implement while at the same timecomplying with established requirements (e.g. each participant should be offered an insertion contract withcertain features within a certain period), with positive or negative financial incentives for those authorities whichdo or do not comply.

73 In order to be constraining for regional authorities, essential levels of provision must be explicitly adopted andcodified by the central level.

74 The autonomy of a territorial public authority from pressures coming from without and its capacity to implementpolicies effectively can be linked to Michael Mann’s discussion of the concept of state power. Elaborating on thisin a way that can be applied to any kind of territorial (and thus sub-national) authority as well, Manndistinguishes “two meanings of state power”: despotic power of the state elite, i.e. “the range of actions which theelite is empowered to undertake without routine, institutionalized negotiation with civil society groups”, andinfrastructural power, i.e. “the capacity of the state actually to penetrate civil society, and to implementlogistically political decisions throughout the realm” (1986:113). Autonomy is therefore related to despoticpower, capacity to infrastructural power.

75 See also Sestito and Nigro (2004).

STEFANO SACCHI AND FRANCESCA BASTAGLI 107

References

Baldini, M., Bosi, P. and Toso, S. (2000) Targeting Welfare in Italy: Old Problems and Perspectives of Reform,Materiali di discussione no. 337, CAPP, Modena: Dipartimento di Economia Politica, Università degli Studi diModena e Reggio Emilia.

Benassi, D. (2000) Le politiche nazionali contro la povertà in Italia, Rapporto per la Commissione d’Indaginesull’Esclusione Sociale, Rome: Poligrafico dello Stato.

Boeri, T. and Perotti, R. (2002) Meno pensioni, più welfare, Bologna: II Mulino.Brandolini, A. (1999) The Distribution of Personal Income in Post-War Italy: Source Description, Data Quality and the

Time Pattern of Income Inequality, Rome: Banca d’Italia.Calza Bini, P., Nicolaus, O. and Turcio, S. (eds) (2003) Reddito minimo di inserimento. Che fare?, Rome: Irpps/

Donzelli.Commissione d’Indagine sull’Esclusione Sociale (2002) Rapporto sulle politiche contro la povertà e l’esclusione

sociale 1997–2001, edited by Chiara Saraceno, Rome: Carocci.Commissione d’Indagine sull’Esclusione Sociale (2003) Rapporto sulle politiche contro la povertà e l’esclusione

sociale 2003, Rome: Poligrafico dello Stato.Council of the European Union (2003) Joint Report of the Commission and the Council on Adequate and Sustainable

Pensions, document of the Council of the European Union no. 7165/03, Brussels, 10 March.Economic Policy Committee (2001) Budgetary Challenges Posed by Ageing Populations: The Impact on Public

Spending of Pensions, Health and Long-Term Care for the Elderly and Possible Indicators of the Long-TermSustainability for Public Finances, EPC/ECFIN/655/01-EN final, Brussels, 24 October.

Esping-Andersen, G. (1999) Social Foundations of Postindustrial Economies, Oxford: Oxford University Press.EUROSTAT (2004) Social Protection in Europe, Statistics in Focus, Population and Social Conditions Theme 3–6/

2004, edited by G.Abramovici, Luxembourg: EUROSTAT.Ferrera, M. (1986) “Italy”, in P. Flora (ed.), Growth to Limits. The Western European Welfare State since World War

II, Berlin: De Gruyter, vol. 2, pp. 385–499.Ferrera, M. (1996) “The southern model of welfare in social Europe”, Journal of European Social Policy, 6 (1):17–37.Ferrera, M. and Gualmini, E. (2004) Rescued by Europe? Social and Labour Market Reforms in Italy from Maastricht

to Berlusconi, Amsterdam: Amsterdam University Press.Ferrera, M., Hemerijck, A. and Rhodes, M. (2000) The Future of Social Europe: Recasting Work and Welfare in the

New Economy, Oeiras: Celta Editora.Ferrera, M., Matsaganis, M. and Sacchi, S. (2002) “Open coordination against poverty: the new EU ‘Social inclusion

process’”, Journal of European Social Policy, 12 (3):227–239.Ferrera, M. and Sacchi, S. (2004) “The open method of coordination and national institutional capabilities: The Italian

experience”, URGE Working Paper no. 2/2004. Available online at http://www.urge.it (accessed 24 May 2004).FIPOSC (2002) Fighting Poverty and Social Exclusion in Southern Europe: Final Report, POLEIS, Milan: Bocconi

University.Gori, C. (2003) “Applicare i livelli essenziali nel sociale”, Prospettive sociali e sanitarie, 15–17:1–8.IRS, Fondazione Zancan, Cles (2001) “Valutazione della sperimentazione dell’istituto del Reddito Minimo di

Inserimento: Principali risultati e indicazioni di sviluppo”, unpublished manuscript.ISAE (2001) “Rapporto trimestrale: Finanza pubblica e redistribuzione, ottobre 2001”, no. 7, Rome. Available online at

http://www.isae.it (accessed 24 May 2004).ISTAT (2003a) “La povertà in Italia nel 2002”, Note Rapide, Anno 8, Numero 2, Rome. Available online at http://

www.istat.it (accessed 24 May 2004).ISTAT (2003b) “L’impatto redistributivo di alcuni provvedimenti contenuti nel disegno di legge finanziaria e nel

decreto legge 269/2003”, Parliamentary Hearing of the President of the National Bureau of Statistics, LuigiBiggeri, Rome, 8 October.

Mann, M. (1986) “The autonomous power of the state: its origins, mechanisms and results”, in J.H.Hall (ed.), States inHistory, Oxford: Blackwell, pp. 109–136.

108 ITALY: MINIMUM INSERTION INCOME

Ministero dell’Economia e delle Finanze (2003) Relazione generale sulla situazione economica del paese (2002), Rome:Poligrafico dello Stato.

Ministero del Lavoro e delle Politiche Sociali (2003a) “Libro bianco sul welfare: Proposte per una società dinamica esolidale”, Rome. Available online at http://www.welfare.gov.it. (accessed 24 May 2004).

Ministero del Lavoro e delle Politiche Sociali (2003b) “Piano di azione nazionale contro la povertà e l’esclusionesociale 2003–2005”, Rome. Available online at http://www.welfare.gov.it (accessed 24 May 2004).

Sacchi, S. (2005) “La lotta alla povertà e all’esclusione sociale nell’Unione europea: il processo inclusione sociale”,forthcoming in M.Ferrera (ed.), Coordinare l’Europa sociale, Bologna: II Mulino.

Sestito, P. and Nigro, V. (2004) “La sensibilità alle regole di accesso della spesa aggregata e della composizione deibeneficiari nel sostegno al reddito di ultima istanza: alcune valutazioni”, unpublished. Available online at http://www. lavoce.info (accessed 24 May 2004).

Toso, S. (ed.) (2000a) Selettività e assistenza sociale, Milan: Franco Angeli.Toso, S. (2000b) “Effetti distributivi della spesa per assistenza in Italia”, in S.Toso (ed.), Selettività e assistenza sociale,

Milan: Franco Angeli, pp. 84–103.

STEFANO SACCHI AND FRANCESCA BASTAGLI 109

4Spain—poverty, social exclusion and “safety nets”

Ana Arriba and Luis Moreno

Introduction

After a long hyper-centralist dictatorship (1939–75), a peaceful transition to democracy (1975–79), and anactive involvement in the process of Europeanization after its accession to the EEC (1986), Spain hasundergone deep and far-reaching social transformations. In economic terms, Spanish development has beenoutstanding: in 1959 the Spanish GDP per head was 58.3 per cent of the EU average; in 1985, 70.6 per centand, by 1998, 81.5 per cent. Spain would match the EU-15 mean by the year 2025 if the annual “catching-up” rate of 0.8 per cent were maintained. No other country in the group of the advanced industrial democracieshas achieved a comparable rate of economic growth. However, economic problems, high levels ofunemployment, a severe demographic imbalance and the abrupt decline of the traditional system ofdomestic care are now threatening the stability of the welfare settlement.

Spain shares with Greece, Italy and Portugal similarities in historical background, value systems, andinstitutional characteristics. The four Southern European countries constitute a distinct welfare regimealongside the wellresearched Anglo-Saxon, Continental and Nordic (Sarasa and Moreno 1995; Ferrera 1996;Rhodes 1996). The role played by the family in the Mediterranean regime constitutes one of its morecharacteristic traits. In households, women have in the past decades deployed a remarkable hyperactivity,which has been crucial for both social cohesion and economic growth.1

Of singular importance regarding Mediterranean welfare development are the manifestations of itscultural-axiological dimension. This is reflected in a self-perception of differentiated needs and lifestyles(intrafamilial pooling of resources, home ownership, and heterogeneity of social reproduction). Alsonoticeable is a compelling household micro-solidarity and a pre-eminence of values of family inclusion andlife-cycle redistribution (gift mechanisms, processes of age emancipation, proliferation of family companiesand jobs). Moreover, cultural choices and practices have structured their civil societies in a characteristicmode (social networking, patronage, clientelism, or group predation).

Contrary to some views regarding Southern European countries as a mere “Latin rim” extra-profitingfrom “social dumping” practices, the evidence show that these countries are the only ones which haveincreased their overall social spending during the 1980s and 1990s (Guillén and Matsaganis 2000). Note thatthe difference between the welfare regime with the highest social spending in 1984 (Continental) and theaverage of the Southern European countries was 9.4 percentage points. This figure was reduced in 1997 to 7.4 as compared to the then highest spender Nordic regime (Castles 2001).

The Spanish welfare state incorporates elements of both Bismarckian and Beveridgean traditions, and canbe labelled a via media with respect to other regimes of social protection. It combines universal and targetedaccess to services and benefits. The welfare state in Spain today represents a fundamental structure for both

social reproduction and political legitimization.2 Since its integration into the European Community in 1986,Spain has followed a pattern of welfare convergence with its European counterparts of a three-fold nature.First, a generalization of social entitlements (education, health, pensions). Second, confluence in the patternof welfare expenditure to the mean of its European partners. Third, a diversification in the provision ofsocial services by private and “third sector” organizations.3

Perhaps the most relevant factor conditioning welfare development in Spain is the importance ofdecentralization both at the level of planning and policy implementation. Spain’s efforts to catch up with theEuropean mean of welfare expenditure have gained impetus with the active role in social policy taken bythe mesogovernments. Often OECD figures have not taken into account the social spending of the Spanishregions, which has dramatically increased their share in the total aggregate of Spanish public expenditure.4

In 1992 Spain’s total social expenditure reached 22.7 per cent of the GDP. However, in 2000 it fell to 20.1,a percentage which compared with the mean figure of 27.3 for the EU-15 (Rodríguez-Cabrero 1994; EU2002). Particularly in social assistance and services areas, decentralization has had a much greater impactthan privatization in the past decades.

As in other Mediterranean countries, the aggregate of Spain’s public policies and interventions associatedwith “safety nets” have developed in a fragmented manner. Indeed, Spanish welfare state expansion hasbeen quite pronounced but it has generally been accomplished inductively on incrementalist bases. As aresult, benefits and services are targeted on different collectives, which often do not have any correlationbetween them, although they sometimes share the same characteristics (Eardley et al. 1996). Note that otherfunctional intervening factors, such as intrafamilial transfers, community help, or altruistic help provided byNGOs (non-governmental organizations) and “third sector” associations, have greatly contributed toconsolidate “safety nets” although much research is needed in order to systematize evidence in this respect.It is worth noting that more than 50 per cent of all the poor in Spain go first to either Caritas or the RedCross in search of support.5

In this chapter, developments in Spanish welfare concerned with policies of poverty alleviation andprogrammes combating social exclusion are analysed. Core areas of observation are the so-called safetynets of social protection, which have been “under construction” in Spain for the past fifteen years.

Safety nets aim at providing citizens and families with basic means that guarantee the satisfaction ofminimum vital needs and facilitate civic integration. Safety nets are the bottom lines of welfare provisionand, as such, they adopt institutionalized expressions that vary in degrees and manners (Moreno 2003). Thecomplex task for the social scientist is to determine the constituent materials from which safety nets areinterwoven. Public programmes of social assistance provided on a means-tested basis are basic constituentelements. In many cases these programmes are of a non-contributory nature, although they may be linked tocontributory social services and subsidies provided by the social insurance system.

In the following section a review of the historical background to the development of social policies inSpain is carried out. The evolution of poverty alleviation programmes is examined within the broaderSpanish welfare state framework. Special reference is made to the creation of regional systems of socialservices and the development of programmes of social assistance by both central and regional layers ofgovernment.

After analyzing the compound nature of the Spanish safety nets, the third section concentrates on thelatest developments and initiatives undertaken with reference to the 2001–03 National Action Plan forSocial Inclusion (NAPSI).6

In the light of the Spanish experience, it is concluded that more attention should be paid to the increasingrole of regional and local layers of government. These, in concurrence with central institutions of EU

SPAIN: POVERTY AND “SAFETY NETS” 111

institutions and of member states, are key actors in promoting the social inclusion of those precariouscitizens facing poverty or social exclusion.

Historical background

The development of welfare in Spain can only be understood in the context of the historical backgroundprior to the transition from Francoism to democracy (1975–78). The peaceful democratic transition wasmade possible by the deployment of consensual politics between representatives, parties and social actors,who accepted the fact that the reform process had to take into account the previous institutional framework.Thus, the consolidation of Spain’s welfare state had to evolve from the institutions and social protectionpolicies developed during Franco’s dictatorship (1939–75).

1939–67: The years of autarchy and the boom of “desarrollismo”

The first period of Franco’s rule was characterized by an attempt to achieve total autarchy with no foreigninterference. Social policy was largely neglected and oriented towards both charity and beneficence. “OurState must be Catholic in the social”, General Franco had already declared in 1937. The Obra Social wasset up in order to “bring joy and bread to the Spanish families”. The worker was regarded as a unit ofeconomic production and, accordingly, was entitled to receive social protection against unexpected risks. Inturn, he or she should be obedient and diligent. A peculiar aspect incorporated in the Fuero del Trabajo, aconstitutional Labour Act implemented by Franco’s Government in 1938, concerned the social status ofwomen. State intervention pursued the “return” of women from the factory to the household and socialprotection applied to only male breadwinner families.

The National Institute for Social Provision (Instituto Nacional de Previsión Social) was the institutionresponsible for social insurance. It incorporated gradually and in a rather disorderly manner theadministration of the various existing benefits and subsidies. In 1938, a Family Subsidy (Subsidio Familiar)was established. In 1942, the programme of Plus Familiar was also created. This subsidy was basically asupplement to salaries. In 1939, an old age subsidy had been introduced to provide benefits for low-wageemployees. This turned into the SOVI in 1947 (Seguro Obligatorio de Vejez e Invalidez, compulsoryinsurance for retirement and invalidity). In 1942, the Statutory Sickness Insurance was also implemented(SOE, Seguro Obligatorio de Enfermedad). Its duration was limited, regardless of the health condition ofthe claimant, and its implicit aim was to “encourage” the employee to go back to work as soon as possible.7

Along with the compulsory social insurance, the old system of mutual insurance was restored. This lattersystem was meant to be complementary to the former, but both systems developed in an overlapping andduplicating manner. Labour mutualidades (mutual aid associations) were non-profit-making associations,organized according to occupational and territorial criteria. They were financed by payroll taxes anddeveloped according to a system of capitalization.

Public intervention for those not engaged in paid labour was paternalistic and provided along charitylines. Social services were very residual and based in the public sphere and were integrated into the localadministration (mainly provincial administrations, or Diputaciones), and under the organic responsibility ofthe Home Ministry at the central level of government. Benefits were distributed by the Social CharitableProtection Fund (Fondo de Protección Benéfico-Social), which had a share of profits accrued by public lottoand games, and which also received private donations. Help and assistance to beneficiaries were provided ina rather discretionary manner in situations of visible need.

112 ANA ARRIBA AND LUIS MORENO

Given the scarce public intervention in programmes of poverty alleviation, subsidiary action was carriedout by private organizations, mainly associated with the Catholic Church, and by foundations such as theRed Cross and some quasi-public Saving Banks (obras sociales de las Cajas de Ahorro). The fascistFalange, the only legal party in Franco’s dictatorship, also developed an institution for social assistance(Obra de Auxilio Social). This was an authoritarian, para-state body, which was based upon the voluntarywork (in many cases forced by compulsory social service for women, along the lines of militaryconscription for men).

The second period of Francoism was characterized by economic “desarrollismo” (developmentalism),which started with the implementation of the Stabilization Plan of 1959. This new economic policy markedthe turning point for the progressive liberalization of the Spanish economy. Technocrats of the Opus Deiheld the key posts in Franco’s government and imported models of “indicative planning” from France.8

Some steps were taken to implement a system of labour regulation that moderated somewhat its dirigistenature. In 1958, for example, a Collective Agreements Act allowed employers and employees to negotiatewages (which had previously been regulated by the Ministry of Labour) at the factory level. In 1963 theMinimum Wage (Salario Mínimo Interprofesional) was legally established. Note that full employment waspractically achieved at the cost of the emigration of many Spanish workers to central European countries.

Social expenditure grew significantly during the 1960s: from 35.5 per cent in 1960 to 55.9 per cent in1970 of the total public spending. Note, however, that the increase was a meagre 3 percentage points of theGDP. Social assistance continued to play a secondary role in the system of social protection and incomemaintenance contributory programmes were the main priority in reforms that extended coverage to affiliatedemployees of the social security system.

In 1963, the Basic Law of Social Security (Ley de Bases de la Seguridad Social) was published althoughits actual implementation was delayed until 1967. It had a universalistic vocation and went hand in handwith a timid fiscal reform. A peculiar model of economic development—vocationally neo-Keynesian butconstrained by the rigidities of an authoritarian regime —attempted the transition from an agrarian societyto a fully industrialized polity with some degree of success, modifying in this process the occupationalstructure of the country.

Outside the contributory realm, there was a very meagre system of social assistance: reduced coverage ofunemployment, absence of welfare cash benefits, insufficient social services with a high degree ofinstitutional dispersion, a prevalence of beneficence criteria and approaches, and paramount concern towardskeeping the social order. Both the Catholic Church, benefiting from large state subsidies, and the family(namely women’s caring activities), played a crucial subsidiary role during Francoism (Arriba 1999).

In 1960, several funds for financing social assistance were legally established, such as the National Fundfor Social Assistance (Fondo Nacional de Asistencia Social, FONAS). This Fund had the aim of supportingcharity institutions and of distributing funds to develop some social services. In the late 1960s, the CatholicChurch reinforced its subsidiary role according to the doctrine of Social Catholicism. This was evident inthose activities of Caritas which pursued a more comprehensive course of action away from thosetraditional charities of mere relief to the needy.

After the 1959 Stabilization Plan, and in the absence of an entrepreneurial class, Franco’s technocratsfrom the state institutions promoted a process of economic modernization. However, public interventionwas mainly geared at avoiding social unrest by means of combining both paternalism and repression(Rodríguez-Cabrero 1989).

SPAIN: POVERTY AND “SAFETY NETS” 113

1968–78: Late Francoism and the transition to democracy

The opposition movement against Francoism became very active in the late 1960s and the early 1970s.These were years of political and social turmoil that constituted the prelude to the transition to democracyafter the death of Franco in November 1975. In 1969, Juan Carlos was appointed the future head of stateand, in 1970, a preferential treaty was signed between Spain and the European Economic Community(EEC).

In 1972 reform of the Social Security, the Financing and Improvement Law (Ley de Financiación yPerfeccionamiento) was implemented with the aim of expanding social protection, especially concerningtemporary labour incapacity, unemployment and old-age pensions. The linkage of workers’ contributions toreal incomes allowed a significant increase of benefits in the contributory system. New special regimes forthe Social Security system (up to twenty) were created and increased the fragmentation of the contributorysystem. Unequal social contributions and payroll taxes paid to the Social Security system favoured the middleclasses and large companies. Despite some improvements in the level of protection, these reformsunderlined the negative level of internal equalization within the contributory system (Cruz Roche et al.1985).

With the implementation of the General Education Law (Ley General de Educación) in 1970 and theGeneral Law of Social Security Law (Ley General de la Seguridad Social) in 1974, the level of publicspending rose significantly. Considerable wage increases also took place between 1974 and 1976. Thesefactors laid the foundations for the subsequent climate of social consensus that contributed to making thepeaceful transition to democracy possible. The main bulk of social spending since the 1970s has beendevoted to retirement pensions and unemployment benefits.

At the end of the Francoist period in Spain, social assistance was characterized by a scarce publicfinancing, a limited degree of social protection, and the important role played by some private or non-profit-making institutions. The small supply of social services was provided by a variety of both public bodies andprivate institutions.

In 1977, the General Directorate for Social Action and Social Services was established within theMinistry of Health and Social Security. This governmental body took over those responsibilities of socialassistance which previously had been attached to various departments of the central government (the HomeMinistry, principally). It ran programmes such as those related to the Social Assistance Fund (Fondo deAsistencia Social), whose non-contributory benefits covered old age and disability pensions outside thesystem of social security. This was considered to be the principal instrument of social assistance at that time.

Within the contributory system of the Social Security, there were also services for legal employees andtheir dependent family members. The system was re-structured with the grouping of old age and disabledservices into the Institute for Social Services (INSERSO Instituto de Servicios Sociales), a quasi-autonomous public agency of the Social Security operating within the framework of the Ministry of Labour.

In parallel, local authorities (municipalities and provincial authorities) continued to run variousprogrammes of social assistance, which were in many cases the inheritors of traditional public charities andbeneficence. Private institutions also continued to provide charitable donations and some services,particularly those offered by Catholic Church organizations.

In the period of transition to democracy, renewed citizens’ demands for the implementation of new socialservices were coupled with an active mobilization of social workers in order to develop a new framework ofservice provision (Sarasa 1993; Casado et al. 1994).

114 ANA ARRIBA AND LUIS MORENO

1978–87: The establishment of the regional systems of social services

A historical review of the reforms implemented in Spain since the inception of the democratic Constitutionof 1978 has to acknowledge the fragmented and inductive nature of the reforms implemented during thetransition to democracy. According to the Spanish Carta Magna, social assistance is a regional power of the“exclusive competence” of the seventeen Autonomous Communities (Art. 148:1.20). Powers concerning thebasic legislation and the economic regime of the Social Security system remained in the hands of the centralgovernment. However, the Comunidades Autónomas (Autonomous Communities) could exercise executivepowers in the running and managing of contributory programmes which could be decentralized to them(Art. 149:1.17).

The constitutional provisions neither defined nor regulated the non-contributory realm of socialassistance and social services. Furthermore, all those powers and responsibilities which were not listed asthe “exclusive competence” of the central government could be requested and exercised by the AutonomousCommunities (Art. 149:3).

As a consequence of the flexibility of the constitutional provisions, all Autonomous Communitiesclaimed in their Statutes of Autonomy (regional constitutional laws) a large number of services andfunctions concerning social assistance, social services, community development, social promotion andwelfare policies in general. The only social services which remained outside the jurisdiction of themesogovernments were those of the INSERSO. However, and as we will analyse below, during the 1990sthe executive powers for the running of practically all INSERSO social services were also transferred to theComunidades Autónomas.

During the period 1982–93, the Autonomous Communities took the legislative initiative in their regionalparliaments, and passed acts which established regional systems of social services. In these pieces oflegislation there was no reference to social assistance as such. The implicit assumption was that, accordingto a comprehensive interpretation, social assistance was an “exclusive” power of the AutonomousCommunities, alongside the social services. In this period the main concern of the Spanishmesogovernments was to request and gain as many powers from the central administration as a flexibleinterpretation of the 1978 Constitution could allow. The Comunidades Autónomas subsequently madeextensive use of their constitutional prerogative for the purposes of institutional legitimization.

In order to develop an integrated network of social services, a common claim in the legislation adoptedby the regional parliaments was the principle of decentralization. According to this, local governmentswould carry out the bulk of service provision,9 but the powers of legislation, planning, and co-ordinationwith the private and altruistic sectors would remain with regional executives and legislatures.

All regional laws envisaged the area of social services as an integrated public system, open to all citizenswithout discrimination. Universal access and equal treatment were, thus, the two ideological foundations ofthe regional legislation on welfare provision. Traditional public beneficence was to be “updated” so thatstigmatization of beneficiaries should be avoided. All things considered, the aim was one of modernizationof the social services in line with other experiences of welfare provision in Western Europe. Such aspirationswere to be in tune with the aim of rationalizing the provision of new services. To this end, inputs of acomprehensive, equal and public nature put forward by a new generation of trained social workers werehighly influential.

The idea of the welfare “mix” was also embraced enthusiastically by most Spanish mesogovernments sothat social intervention could be optimized. Non-profit-making organizations, in particular, wereincorporated in the general provision of social services, and many of them were subsidized by the regionalpublic bodies.

SPAIN: POVERTY AND “SAFETY NETS” 115

The processes for the implementation of regional systems of social services in Spain were not developedwithout a degree of friction with the central government. In 1986, a decision of the Constitutional Court(146/1986) corroborated the principle that, despite the “exclusive” powers of the Autonomous Communitiesin this field, the central government could also develop programmes of social assistance guaranteeing anequal treatment to all Comunidades Autónomas.

Among the initiatives taken by the central government during this period, the passing of the LISMI (Leyde Integración Social de los Minusválidos, Law for Social Integration of the Handicapped) Law (1982) is tobe underlined. This piece of legislation provided guaranteed benefits to citizens with a disability (65 per centor higher). In 1983–84, the Socialist PSOE Government also attempted unsuccessfully to enact a NationalAct of Social Services. This law would aim at integrating all “scattered” social services within thecontributory system of the Social Security into one institutional framework, centrally managed. But theSpanish mesogovernments remained as the main protagonists in the area of welfare development, a logicalconsequence of the home-rule-all-round process of decentralization.

1988–2000: The expansion and consolidation of welfare programmes

The year 1988 can be identified as the beginning of a cycle of major developments concerning socialservices in Spain, and of a big expansion of social spending. The establishment in 1988 of the Ministry ofSocial Affairs10 aimed at the development of a social policy, an area of public intervention which had been“hidden” within the organizational structure of the Ministry of Labour. However, central intervention wassomewhat conditioned by the consolidation of the regional systems promoted by the Spanish ComunidadesAutónomas. Some views were expressed against the establishment of a Ministry which would be lackingmany powers already decentralized to the mesogovernments (Beltrán 1992).

But the newly created Ministry of Social Affairs pursued a course of action of co-ordination with themesogovernments in the development of General Plans such as those concerning old age, drug addiction,equal opportunities, or youth. These Plans were not passed as pieces of legislation by the Spanish Parliament.They were agreements aimed at making functional the structures of welfare provision in the whole of Spain.In particular, they paved the way for the future transfer of the social services of the INSERSO to theregional systems of social services.

The most important agreement between the three layers of governments had already taken place in 1987with the approval of the “Concerted Plan for the Development of the Basic Provision of Social Services bythe Local Authorities” (Plan Concertado para el Desarrollo de Prestaciones Básicas de Servicios Socialesde las Corporaciones Locales). This intergovernmental agreement has resulted in administrative co-operation between central, regional and local governments. The aim is to provide services at the municipallevel for the following purposes: (1) information and counselling; (2) social and day care services for thedisabled and elderly; (3) refuge for abused women, single mothers, orphans or mistreated minors, andsheltered housing for the homeless; and (4) prevention and social insertion.

This network of centres constitutes the basic level for primary attention in Spain, and was supported byall Autonomous Communities except the Basque Country.11 The annual financing of this Plan is met onequitable terms by the three layers of governments. This agreement was the first in a model ofintergovernmental relations characteristic of the process of federalization of politics in Spain. Itsimplications for other policy areas have been of no little significance (Agranoff 1993; Moreno 2001b).

Already in 1987, Spain had joined the EEC II Programme of Fight Against Poverty. This involvementenabled the collaboration of various private and public institutions at the different levels of government.

116 ANA ARRIBA AND LUIS MORENO

Comparisons with the situation in other European countries were drawn. As a result, growing concern aboutpoverty was noticeable at the various institutional levels and cognitive domains (intervention and research).

During this period the major reform carried out by the central government concerning low-incomecitizens was the universalization of old age and disability pensions. These pensions have become afundamental component in the Spanish safety net. In 1990, the Law of NonContributory Pensions of theSocial Security (26/1990) put into effect the awarding of means-tested benefits for old age and disabledcitizens non-affiliated to the Social Security system, or with discontinuous contributory records. Note thatthe financing of these benefits was initially met by the contributory Social Security system, although it wasagreed later on by the social partners in the Toledo Pact (see below) that the costs of non-contributorypensions would be charged to the Spanish general revenue. The 26/1990 Law also established the awardingof means-tested benefits to low-income families for children under 18 years and handicapped, and to whichboth families in the contributory and non-contributory systems were entitled to claim.

The most substantial developments regarding social assistance and non-contributory benefits wereimplemented during this period. All things considered, the single most innovative contribution in theconstruction of safety nets in Spain has been made possible by the process of decentralization of power tothe seventeen Comunidades Autónomas. These have developed an active role regarding the implementationof policies directed to fight poverty and social exclusion.

In parallel, and mostly during the 1990s, the regions decided to implement their programmes of minimumincome for insertion (Rentas Mínimas de Inserción). These are programmes of minimum income guaranteedfor low-income families, which aim at facilitating social insertion of the recipient families (along the linesof the early principles established by the French RMI). These minimum income benefits filled in a hole inthe social protection system or, in other words, helped to complete the weaving of the Spanish safety net asit stands now. Such benefits were intended to provide monetary resources to those citizens potentially activein the labour market facing situations of need. Elaboration and implementation of these programmes by theSpanish regions took place between 1989 (first programme introduced in the Basque Country) and 1995(final one implemented in the Balearic Islands). Some main features of these programmes will be analysedin this chapter. Let us point out now that, despite their common initial purposes, they are diverse as regardsprotection intensity, coverage or means to achieve social integration of programme beneficiaries.

After the victory of the Popular Party (PP) in the 1996 General Elections, the Ministry of Social Affairswas subsumed into the Ministry of Labour and Social Affairs. The former Ministry’s powers andcompetencies were integrated in the lower politically ranked Secretar�a General de Asuntos Sociaks. Alsoin 1996 the INSERSO changed its name with the assumption of competencies concerning migration. It wasrenamed IMSERSO (Instituto de Migraciones y Servicios Sociales—Institute for Migration and SocialServices). Since then, it has kept minor executive programmes regarding services for immigrants, as well assome functions of co-ordination with EU programmes. Let us remember that the executive responsibilitiesfor the running of social services of the former INSERSO had already been handed over to the AutonomousCommunities, a process completed in 1996.

The considerable expansion of both regional social services and social assistance programmes hasresulted in a de iure division between the contributory (social insurance or social security system for“insiders” and legal employees) and social assistance realms of welfare provision (for those excluded fromthe formal labour market). However, both domains are intertwined in an aggregate of social provision. Thishas been inspired by the general principle of social citizenship and has, thus, expanded the “grey zones”between both social insurance and welfare assistance realms (Moreno and Sarasa 1993).

Note that in the period 1980–92, the total number of pensioners rose by 2.5 million (2.1, contributory, and0.4, non-contributory), from 4.7 to 7.2 million. The total expenditure increased from 5.9 per cent (5.8 per

SPAIN: POVERTY AND “SAFETY NETS” 117

cent, contributory, and 0.1 per cent, non-contributory) to 8.6 per cent (8.1 contributory, and 0.5 non-contributory) as a percentage of the Spanish GDP. Average social security contributory pension benefitsincreased from 66.5 per cent of the legal minimum wage, in 1980, to 93.3 per cent in 1992, and 100 percent, in 1995. Non-contributory pensions amounted to 53.3 per cent of the legal minimum wage in 1992(Cruz Roche 1994).12

In the late 1990s reforms have concentrated on the consolidation of the general system of socialprotection, especially as concerns financing. This course of action was based on the consent of the mainparliamentaiy parties and social agents (trade unions, principally). The Report on the “Proposal for theAnalysis of the Structural Problems of the Social Security System and of the Main Reforms to beAccomplished” (Ponencia para el Análisis de los Problemas Estructurales del Sistema de la SeguridadSocial y de las Principales Reformas que deberán acometerse) was approved by the Congress of Deputies(Spanish parliamentary Lower House) on 6 April 1995, and has since then been known as the Pacto deToledo.13 Among the main objectives pursued, two in particular can be emphasized:

1 A clear division between contributory social insurance and universal non-contributory benefits. Thisimplies that benefits and services of a universal nature (health and social services), and assistancesubsidies (pensions’ supplements, non-contributory pensions and cash benefits for dependent children)are to be financed by transfers from general taxation revenue. Contributory benefits, in turn, are to be metby both employers’ and employees’ social contributions.14

2 Pensions (contributory and non-contributory) are to be graded annually in line with price increases (andretroactively).

Another main line of reforms has been the gradual orientation towards the activation of policies targeted atclaimants potentially active in the labour market. This has been coupled with the adoption of a somewhatrestrictive criterion of access to benefits and unemployment subsidies. Access to the latter has beenrestricted even further regarding previous contributory requirements and benefit duration.15 Insertion goalsincluded in the regional programmes of minimum income have been the most visible orientation of thisactivation approach. However, this workfare development has also extended to unemployment subsidies(e.g. active income for insertion, requirements for both agrarian unemployment and general unemploymentsubsidies, or schemes for “flexible” retirement). Furthermore, programmes targeted at favouring access tothe labour market are also taking into account incentives for workfare activation of those sectors ofpopulation that are socially excluded.16 This approach is meant to optimize both the policies of labouractivation by employment services and those undertaken by social services for the purposes of socialinsertion.

Some policies for a limited family support have been also implemented in recent years. An instance ofthis is the Individual and Family Minimum benefit (Mínimo Personal y Familiar) according to the 1998Income Tax Act (40/98). This Mínimo, or non-taxable minimum, was defined as the disposable income ofthe taxpayer to comply with his/her basic needs and with those of his/her dependent family members (seeTable 4.3).17

Other measures have been aimed at facilitating conciliation between family and labour lives (Ley bor laConciliación de la Vida Laboral y Familiar, 39/1999), and some of them implicitly favour a pro-natalistpolicy.18 According to Decree Law (1/2000) the already low benefits for family protection were slightlyincreased.19 Likewise, new cash benefits were to be awarded for the birth of a third and successive child, aswell as in the case of multiple births (they are small one-off payments as reproduced in Table 4.3).

118 ANA ARRIBA AND LUIS MORENO

The process of decentralization of policies and services has been deepened and reinforced in recent times.However, the principle of decentralization put forward by the regional legislation has often brought about acertain re-centralization of the policy-making process back to the intermediate layer of government (i.e.regional). Important political decisions regarding the minimum income schemes, and the organization andplanning of the services developed according to the Plan Concertado, have reflected not only a higherdegree of political dynamism by the Comunidades Autónomas. They have also underlined the subsidiaryrole played by local councils, dependent to a great extent on regional financial sources and regional politicalconcerns.20 Only the big cities have been able to challenge the mesogovernments as main protagonists in thedevelopment of social services.

The social situation: recent developments

Poverty trends

Prior to analyzing some of the most relevant features of poverty in Spain, it is necessary to point out thedifficulties in obtaining data from official sources. Unlike the situation in other European countries, in Spainthere are no official statistics on socio-economic inequalities and poverty. As a consequence, researchers onthese issues must deal with other “indirect” sources of information (Ruíz-Huerta and Martínez 2000). Inrecent times there have been three main sources for data collection on poverty:

• The Spanish Surveys on Family Accounts (Encuestas de Presupuestos Familiares, EPF). Since the1970s, these have been carried out by the Spanish National Institute of Statistics (INE) in order toprovide data on Spaniards’ consumption patterns. They have been the main source of empirical evidencefor studies on inequality and poverty analysis and for comparisons made with other countries. EPFinformation deals mainly with expenditure consumption patterns, as well as monetary income, data onhousehold equipment and appliances, housing features and data on the subjective perception of thesurveyed of their economic situation.21

• The European Community Household Panel, ECHP. Co-ordinated by EUROSTAT, this aims atproviding comparable data for the study of the living conditions in European households (including thestudy of social inequalities, poverty and social exclusion). Because of the lack of other national officialsources, the ECHP has become a crucial resource for poverty research in Spain.22

• Intermittent and one-off studies promoted by private organizations, among which the most relevant oneshave been those carried out by Caritas and Edis-Caritas.23 Others have dealt with sub-nationalframeworks (sponsored by the municipal, provincial and regional tiers of government), and have oftenbeen carried out concerning the implementation or evaluation of programmes of poverty reduction.

In this section we take into account data on poverty produced by the first two main sources (EPF-ECPF andECHP). Let us remember that data provided by the successive findings of the European CommunityHousehold Panel (ECHP) on household income are detailed and allow longitudinal studies and Europeancross-country comparisons. As ECHP data do not provide territorially disaggregated information at theregional level, we make use of the 1998 ECPF data instead, which offer a regional breakdown of figures.We take into consideration the most recently available data concerning both sources of information.24

The first observation on the figures provided by the various sources is the wide disparity concerning theSpanish poor population according to different indicators (see data reproduced by the MTAS, Spanish

SPAIN: POVERTY AND “SAFETY NETS” 119

Ministry of Labour and Social Affairs, in the 2001 NAPSI document, Plan Nacional para la InclusionSocial).

Figure 4.1 Poverty rates in Spain (ECHP 1994–97) (source: Ayala et al. (2002)).NotePoverty rates correspond to percentage shares of population (individuals) whose monthly equivalized income is atdifferent thresholds. Data from the European Community Household Panel (OECD modified scale).

In spite of the disparity of the available data on poverty, the various methodologies and research carriedout until now allow us to point out consistent trends and figures. During the 1960s and 1970s, analyses onpoverty were generally carried out within more general research studies on social structure and change inSpain.25 The very few studies carried out during the 1960s described poverty in a country on the brink ofindustrialization facing problems of infrastructure, housing and developmental services (Ayala and Renes1998). Poverty in the 1970s was analysed in studies published mainly during the 1980s and the beginning ofthe 1990s (the latter based upon data provided by EPF surveys),26 which coincided with the process ofpolitical transition to democracy and reflected a decrease in the levels of relative poverty. After theinternational economic crisis in the later part of the 1970s there was an increase in the poverty rates aroundthe beginning of the 1980s. In the subsequent period of expansion the number of poor households decreasedand the balance at the end of this decade saw a reduction in the rates of relative poverty (Cantó et al. 2002).

In the mid-1980s, and on the request of Caritas, the study on “poverty and marginalization” carried out byEDIS (1984) had a considerable impact not only on the media, but also provoked a wide debate in thepublic opinion. The study estimated that around 8 million Spaniards were poor (i.e. having less than 50 percent of the income mean). The findings of the EDIS study allowed Caritas to initiate a campaign ofdemands to the government and public authorities. Following this debate, the number of studies on povertyincreased considerably (Arriba 1999; Susín 2000).

During the 1990s (see Figure 4.1), and taking into consideration ECHP data from 1994 to 1997, the rateof relative poverty (both at 50 per cent of the mean and 60 per cent of the median equivalized income) wasbetween 19 and 20 per cent of the whole Spanish population (Ayala et al. 2002). The oscillation of the ratesof absolute poverty (25 per cent of the mean and 30 per cent of the median of equivalized income) wassomewhat higher: between 3.5 and 6.6 per cent of the population.27

ECHP data has made it possible to research the fluctuation of the static and the dynamic types of poverty.According to the findings of García Serrano et al. (2001), permanent poverty in Spain was 9.8 per cent in

120 ANA ARRIBA AND LUIS MORENO

the period 1994–96, while 15.1 per cent of the population was entering and leaving situations of poverty;the rest (75.1 per cent) was not poor in either of the cases under consideration.

Table 4.1 reproduces information about socio-demographic characteristics of individuals (over 16 years ofage) living below the national threshold of relative poverty (50 per cent of mean income) based on ECHP1996 data. In the first column, there were percentages of poor grouped according to individual andhousehold features. According to age, the group with higher poverty rates is the youngest (24.4 per cent for16–24 years), followed by that of 45–54 years (20.1 per cent). It is worth noting that the lowest rate ofpoverty is among people over 65 years (10.2 per cent).

Table 4.1 Poverty rates by individual and household characteristics ECHP 1996 (%)

Poverty risk by groups Poor population (distributionby individual and householdcharacteristics)

Total population (distributionby individual and householdcharacteristics)

Age of individual

16–24 24.4 22.9 16.5

25–34 16.6 18.2 19.2

35–44 17.5 17.3 17.2

45–54 20.1 16.2 14.1

55–64 18.4 13.8 13.2

� 65 10.2 11.6 19.8

Educational level

High 5.9 4.7 14.1

Middle 12.3 13.1 18.6

Low 21.4 82.1 67.2

Marital status

Married 17.0 58.1 60.1

Separated/Divorced 24.9 2.5 1.7

Window 10.6 4.9 8.1

Single 20.1 34.5 30.1

Household size

1 4.2 1.2 4.9

2 10.6 10.2 16.9

3 14.7 18.3 21.9

4 16.2 26.9 29.0

5 26.4 25.5 16.9

� 6 30.6 17.9 19.4

Household type

Single 16–64 7.9 0.7 1.5

Single � 65 2.7 0.5 3.5

Couple no children � 65 13.8 5.0 6.3

Couple no children � 65 7.9 4.7 10.3

Other no children 15.4 21.7 24.6

SPAIN: POVERTY AND “SAFETY NETS” 121

Poverty risk by groups Poor population (distributionby individual and householdcharacteristics)

Total population (distributionby individual and householdcharacteristics)

Single with children 24.7 1.2 0.9

Couple 1–2 children 14.5 18.5 22.5

Couple children 32.2 6.8 3.7

Other with dependent children 27.0 35.7 23.2

Labour market position

Working 11.7 26.0 39.1

Unemployed 38.3 24.2 11.1

Inactive 17.5 49.8 49.8

Poverty risk by groups Poor population (distribution byindividual and householdcharacteristics)

Total population (distribution byindividual and householdcharacteristics)

Main household incomes 1995

Wage and salary 13.3 41.5 54.6

Self-employed 21.7 13.4 10.8

Pension 14.3 19.4 23.7

Unemployment subsidy 64.4 10.4 2.8

Other social benefits 36.6 11.2 5.3

Capital income 20.7 1.9 1.6

Total population 17.5 100.0 100.0

Sources: Elaboration from García Serrano et al. (2001).NotePoverty threshold: 50 per cent mean of monthly equivalized income (OCDE standard scale). Data ECHP(1996).

This finding is consistent with the poverty rates among people over 65 years living alone (2.7 per cent), incouples (7.9 per cent), or among retired pensioners28 (7.9 per cent). Taking into account the sources ofincome, poverty rates were much higher in those groups of individuals living in households where theirmain means of income were unemployment subsidies (64.4 per cent), followed by those receiving othertype of social benefits (36.6 per cent). Access to a salary or a contributory pension is, according to thesedata, the best protection against the risk of poverty in Spain. Poverty rates were higher among unemployed(38.3 per cent) and individuals with low levels of formal education (21.4 per cent). According to legalstatus, poverty rates are also higher among widowers or divorcees (24.9 per cent) and among singles (20.1per cent). As regards types of household, rates were higher among larger ones (26.4 per cent in householdsof five members, and 30.6 per cent in households of six or more members) and those with dependentchildren (32.2 per cent in households with three dependent children and 24.7 per cent in those with onesingle parent living with dependent children).29 At the end of the 1990s, it could be pointed out in general termsthat the youngest, the single unemployed, households with unemployed and people living in large-sizehouseholds are the ones facing greater risks of being in situations of poverty.30

As regards regional poverty, data on the territorial distribution of poverty reflect a traditional divisionbetween northern and southern regions, although this cleavage does not reach the abrupt proportions of theItalian case. As shown in Table 4.2, Extremadura, Murcia, Canaries

122 ANA ARRIBA AND LUIS MORENO

Table 4.2 Territorial distribution of poverty in Spain (ECPF 1998)

Autonomouscommunities

Share of totalpopulation

Poverty rates Share of total poorpopulation

Poverty gap

Andalusia 18.3 20.7 31.4 4.8

Aragon 3.0 9.3 2.3 1.6

Asturias 2.7 6.9 1.5 1.4

Balearics 1.9 3.6 0.6 0.4

Basque Country 5.2 8.5 3.7 1.8

Canaries 4.0 18.3 6.1 3.8

Cantabria 1.3 13.0 1.4 2.6

Castile-La Mancha 4.3 13.0 4.7 2.6

Castile and Leon 6.3 12.6 6.6 2.1

Catalonia 15.4 5.8 7.4 0.9

Extremadura 2.7 22.6 5.2 4.8

Galicia 6.9 15.2 8.7 3.3

La Rioja 0.7 3.2 0.2 0.4

Madrid 12.8 7.2 7.6 1.6

Murcia 2.8 22.0 5.1 4.1

Navarre 1.3 4.6 0.5 0.8

Valencia 10.0 8.0 6.7 1.6

Ceuta and Melilla 0.3 14.9 0.4 4.7

Total 100.0 12.1 100.0 2.5

Sources: Elaboration from Ayala (2002). Data: Spanish Survey on Family Accounts, ECPF, 1997.NotePoverty rates refer to percentage shares of population whose monthly equivalized income is under regional thresholds

(50 per cent mean of monthly equivalized income) (modified OECD scale).

and Andalusia have poverty rates (measured as 50 per cent of the mean equivalized income at the regionallevel) over 18 per cent of the regional population. Note that poor population in these regions almost reachhalf of the total poor population in Spain (or 47.8 per cent).

Above the Spanish national poverty rate there is a group of regions with lower levels of economicdevelopment (both Castilles, Galicia or Cantabria). Regions with industrial growth poles and fasterdevelopment of the service sector of the economy have poverty rates lower that 10 per cent (Aragon,Asturias, the Basque Country, Catalonia, Valencia and Madrid are between the 5–10 per cent). In theBalearic Islands, Navarre and La Rioja, the regions highest levels of regional income, poverty rates hardlyreach 5 per cent. The poverty gap31 is a good index for comparing the various degrees of severity in thedifferent regions. But the results after applying this index do not result in a different situation from thatalready reflected by the regional poverty rates. Thus, in those regions where there are more poor people, thepoverty gap also shows a higher degree of severity.

Nevertheless, such indexes do not incorporate the dynamics in the processes of narrowing or wideningthe disparities among the diverse regions or Comunidades Autónomas concerning poverty levels. At thispoint, it is important to assess the impacts that the general process of political decentralization and thepolicy-making implemented by the Spanish mesogovernments have had on poverty reduction. Diverseestimates carried out on the evolution of recent regional and individual economic disparities in Spain have

SPAIN: POVERTY AND “SAFETY NETS” 123

confirmed that a reduction of inequalities concerning the per capita family income among ComunidadesAutónomas has taken place during the 1990s. Potentialities for policy innovation by the regions have not ledinevitably to an increase in territorial inequalities. So far the “demonstration effect” between the regions hasproved to be a great “equalizer” in terms of general welfare policy output. Evidence has shown thatterritorial unbalances account for only around 10 per cent of the personal inequalities and had alreadytended to decrease during the 1980s. Personal redistribution produced by the impact of direct taxation, socialcontributions and monetary transfers had significantly reduced regional disparities in Spain by between 25and 34 per cent (Moreno 2002b).

These data provide a concise overview of poverty figures in Spain, but they offer a rather distortedsituation. Beyond the establishment of static thresholds for the calculation of the various definitions ofpoverty, one consideration that is often missed in the analyses of poverty is the importance of family andhouseholds in people’s attitudes, expectations and decisions. In Southern European countries, severepoverty rates are lower due to the role of the family as a “clearing house” in the distribution of materialresources. Note that, individually considered, 36 per cent of the total Spanish population was extremelypoor in 1993 (less than 25 per cent of the mean equivalent household income per head), but they amountedto only 5 per cent of the Spanish population if they were considered to share their income within households(Carabaña and Salido 2001).

Low-income benefits

A look at Tables 4.3 and 4.4 illustrates the various types of benefits and subsidies (both contributory andnon-contributory) which composed the public safety net in Spain. Benefits for low-income citizens can begrouped as follows.

Supplementary Social Security Benefits32

• Social Security Minimum Pension Supplements (Complementos de Mínimos de Pensiones de laSeguridad Social). These are benefits whose function is to top up already provided pensions to aminimum threshold that is legally established. Claimants’ monthly income (excluding the amountreceived as pension) must not be higher than a limit fixed annually. The Minimum Supplements reachlittle more than 30 per cent of the total contributory pensions (2,402,321 in the year 2000). In recentyears the number of supplemented pensions has gradually decreased due to the longer contributorybiographies of the new pensioners. They mostly relate to survivors’ pensions, as well as those in theAgrarian and Domestic Regimes of the Social Security.

• Social Assistance Benefits for the Unemployed (Subsidios de Desempleo): (i) the means-tested Subsidiopor Desempleo is available to those unemployed whose contributory benefit has finished. They areespecially targeted at those with dependent children, or those over 52 years of age with difficulties in re-entering the labour market. During 2000, the mean figure of monthly beneficiaries was 367,851; (ii) theSubsidio de Desempleo Agrario is available to unemployed workers included the Special AgrarianRegime within the Social Security system, and living in the poorer regions of Andalusia andExtremadura (224,170 monthly beneficiaries),33 and (iii) the Renta Activa de Inserción Laboral, abenefit introduced in 2000 on an annual basis, which aims at facilitating labour insertion for long-termunemployed workers over 45 years of age, and who are in needy situations subject to means testing. Thisrentas activas had a limited impact in 2000, when they were implemented (3,966 beneficiaries).

124 ANA ARRIBA AND LUIS MORENO

However, it meant the adoption of an insertion criterion for unemployment subsidies similar to that ofthe regional minimum income schemes.

Non-contributory Social Security benefits

• Non-contributory Social Security Pensions (Pensiones no Contributivas de la Seguridad Social, or PNCs),Assistance Pensions (Pensiones Asistenciales), and LISMI subsidies for the elderly and disabled(Subsidio de Garantía de Mínimos). (NB. The latter is gradually being phased out). The non-contributorypensions are benefits targeted at households of low-income population of +65 years and the disabled(with a 65 per cent of invalidity). Beneficiaries have not been contributory members of the SocialSecurity system during their working life. During 2000, the monthly mean figure of non-contributorypensioners was 471,275. Note that both Assistance and LISMI subsidies are subject to a gradual phasingout after the implementation of the PNCs in the year 1990 (Law 26/1990). They are still available for thosebeneficiaries who had access to LISMI before 1990 and were not included in the provisions of the PNCs.However, during 2000 there were 68,058 and 83,471 Assistance and LISMI pensioners, respectively,whose benefits would be lower if they were to receive the later implemented PNCs.

• Family Benefits of the Social Security (Prestaciones Familiares de la Seguridad Social). These are lowintensity benefits for dependent

Table 4.3 Safety net benefits and minimum schemes: amounts and beneficiaries (2000)

Benefits Amounts % M.W. Beneficiaries Notes

Contributory access

Minimum contributorypensions supplements

With spouse

Retirement 65 years: 424.61 99.3 2,402,321 Beneficiaries up to 31/12/2000: 31.4% ofpensions

Disability: MajorInvalidity:

639.92 149.6

Absolute: 424.61 99.3 All amounts in 14payments per year(with or withoutdependent spouse)

Total with 65 years: 424.61 99.3

Absolute orphan: 353.75 82.7

Without spouse

Retirement 65 years: 360.24 84.2

Disability: Major: 540.82 126.4

Absolute: 360.54 84.3

Total with 65 years: 360.54 84.3

Widows 65 years: 360.54 84.3

60–64 years: 316.94 74.1

>60 years: 252.90 59.1

Orphans: 104.30 24.4

SPAIN: POVERTY AND “SAFETY NETS” 125

Benefits Amounts % M.W. Beneficiaries Notes

To relatives in charge: 104.30 24.4

With 65 years: 268.59 62.8

� 65 years 252.90 59.1

Family protectionsupplements

<18 years

No disability andyounger

than 18 years: 24.25 4.9

Disability

Younger than 18 years

and disability >33%: 48.47 9.8

Older than 18 years

and disability >65%: 234.76 47.4

Benefits Amounts % M.W. Beneficiaries Notes

Older than 18years anddisability >75%:

352.13 71.1

MaximumIncome forfamilyprotection:

619.99 125.1

Unemploymentsubsidy

318.60 64.3 367,851 12 payments peryear

Agrarianunemploymentsubsidy(Andalusia andExtremadura)

318.60 64.3 224,170 12 payments peryear

INEM Activeincome

318.60 64.3 3,966 Maximum: 10payments

Non-contributory access

Assistancepensions(phasing out)

174.83 35.3 68,058 “Frozen”amount since1991 (phasingout)

Beneficiaries:data from theBasque Countryunavailable

Minimumincomeguarantee(disability-LISMI)(phasing out)

174.83 35.3 83,471 “Frozen”amount since1991 (phasingout)

126 ANA ARRIBA AND LUIS MORENO

Benefits Amounts % M.W. Beneficiaries Notes

Beneficiaries:data from theBasque Countryand Navarreunavailable

Non-contributorypensions fordisabled andold-age

288.10 58.1 471,275 12 payments peryear

Regionalminimumincome schemesof theAutonomousCommunities

Maximum: 318.60 64.3 78,445* Basic amount.12 payments peryear. *Numberof households

Minimum: 238.92 48.2

Other minimum schemes

Minimum wage (M.W.) (month) 427.79 Annual amount correspondsto 14 monthly payments

(per year) 5,947.13

Income tax: Individual minimum

Personal/family minimum Taxpayer <65 years 3,305.56 Established by Income TaxLaw 40/1998

(“mínima vital”) � 65 years 3,906.57

Disability: 33–64% 5,108.60

Disability � 65% 6,911.63

Single-parent family <65years

5,409.10

Single-parent family � 65years 65 years

6,010.12

+ disability between 33–64% 7,212.14

+ disability � 65% 9,015.18

Family minimum

Ascendant � 65-y. + income< M.W.

601.01 Established by Income TaxLaw 40/1998

Descendant.

Single < 25 years.*: *Provided that income isnot higher than theminimum thresholdestablished of 6.010.12exempted the amountsregulated in Art. 48 of theIncome Tax normative

1st and 2nd children 1,202.02

SPAIN: POVERTY AND “SAFETY NETS” 127

3rd and successive children 1,803.03

suppl. school material (3–16years)

150,25

suppl for each descendant <3years

300.50

suppl. disability 33–64% 1,803.03

suppl. disability � 65% 3,606.07

Sources: Elaboration from CES (2001) p. 66, MTAS (2001b) and data provided by the Direction General deAction Social, del Menor y de la Familia (Spanish Ministry of Labour and Social Affairs).

Table 4.4 Main features of safety net benefits (2000)

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

ContributoryAccess

Minimumpensionssupplements

Contributory

Onindividualincome

Indefinite

INSS(SocialSecurityNationalInstitute)(Centrally run)

– Differentialamountup to theminimumpension(yearlyfixedaccording to age,familysituationand kindofpension)

Pensionerswhosebasicpensionis lowerthan theminimumIncomelowerthan themaximumamount(fixedyearly)

Familyprotectionsupplements < 18years

Contributory

Onhousehold income

Indefinite whileaccessconditions

INSS – Totalamount

Affiliatedworkersandpensioners

Childwith nodisability andyoungerthan 18years

Incomelower

128 ANA ARRIBA AND LUIS MORENO

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

than thefixedamountyearly(+15%for the2ndchildandsuccessive)

Reducedamount(differential)

Affiliatedworkersandpensioners

Childwith nodisability andyoungerthan 18years

Incomelowerthan themaximumamount(yearlyfixed)+benefit

Disability

Contributory

When noaccess toa non-contributorypension

Indefinite whileaccessconditionsremain

INSS – Totalamount

Affiliatedworkersandpensioners

Disabledchildrenwithoutaccess tonon-contributorypensionsor

SPAIN: POVERTY AND “SAFETY NETS” 129

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

assistancepensions

Unemploymentsubsidy

Contributory

Onhouseholdincome

Variabledepending onthecontribution, ageandfamilycharges

INEM(EmploymentNationalInstitute)(Managementtransferred toregions)

Totalamount

Unemployedpeople(1monthregistered) Donotrefuseright joboffersneithertrainingorlabourrestructuringprogrammes

Exhaustedbenefitorinsufficient

contribution andothersituations

LowincomesContribution: 3months+familyresp.; 6months;+52yearswithaccessto cont.pension

130 ANA ARRIBA AND LUIS MORENO

+6 yearsofcontributions

Olderthan 45withexhausted benefit

Agrarianunemploymentsubsidy(Andalusia andExtremadura)

Contributory

Onhouseholdincome

Variabledepending onthecontribution, ageandfamilycharges

INEM Conditions: Itdependson ages(16, 25,52 and60years),access,amountsandlength

Totalamount

UnemployedpeopleEventualemployeeregistered in theREASS,Andalusia andExtremaduraLowincomesContribution:minimumnumberofworkingdays(35);+52yearswithaccessto cont.pension+6 yearsofcontributions

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

INEMactiveincome

Contributory

Onhouseho

10months

INEM-CCAA(Region

Labourinsertion

Totalamount

Long-termunemplo

SPAIN: POVERTY AND “SAFETY NETS” 131

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

ldincome

alAutonomousCommunities)

programme andworkfarecommitment

yedpeople(12months)� 45yearsFamilyresponsibilityLowincomesEnd ofthecontributorybenefitSigningofworkfarecommitment(benefitisprovided3monthslater)

Non-contributoryaccess

Assistancepensions(phasingout)

Non-contributory

Onhouseholdincome

Indefinite

CCAA – Totalamount

Disabledand old-age � 66yearsHelplessnessIncomelowerthan thebenefitand noproperty

Minimumincomeguarantee(disabilit

Non-contributory

Onhouseholdincome

Indefinite

IMSERSO-CCAA(Institute ofMigratio

– Differentialamount

DisabilityPersonalincomelowerthan

132 ANA ARRIBA AND LUIS MORENO

Benefits Access Meanstesting

Length Administrativeagency

Activationmeasures

Estimated amount

Requirements forbeneficiaries

y-LISMI)(phasingout)

ns andSocialServices)(Managementtransferred toregions)

70%MW

Non-contributorypensionsfordisabledlow andold-age

Non-contributory

Onhouseholdincome

Indefinite

IMSERSO-CCAA

– Differentialamountbetweenincomesorbeneficiarybenefits/household andthe 2ndbeneficiarypensionandsuccessive: 70%

Invalidity: 18–64years,� 65%disability, lowincomes

Minimum: 25%of thepension

Retirement: 65years,insufficientincomes

Non-contributoryfamilyprotection < 18years

Non-contributory

Onhouseholdincome

Indefinite whileaccessconditions exist

INSS Totalamount

Childwith nodisability andyoungerthan 18years

– Incomelowerthan thefixedamountyearly(+15%

SPAIN: POVERTY AND “SAFETY NETS” 133

for the2ndchildandsuccessive)

Reducedamount(differential)

Childwith nodisability andyoungerthan 18years

Incomelowerthan themaximumamount(yearlyfixed)+benefit

Disability

Non-contributory

Whennoaccess toa non-contributorypension

Indefinite whileaccessconditions exist

INSS – Totalamount

Affiliatedworkersandpensioners

Disabledchildrenwithoutaccessto non-contributorypensionsorassistancepensions

Minimumincomeschemesof theAutonomousCommunities

Non-contributory

Onhouseholdincome

Variabledepending ontheautonomouslegislation

CCAA Commitmentsandsocialandlabourinsertionprogrammes

Differentialamount

Variable

134 ANA ARRIBA AND LUIS MORENO

children under 18 years awarded to low-income families either with or without contributory biographies(see Table 4.3). They are slightly more generous in the case of handicapped children. Benefits amountsare higher in the case of children over 18 years with serious disability, and in this respect they can beconsidered as non-contributory pensions although the benefit is incompatible with any other non-contributory benefit. During the year 2000, amounts were increased and other benefits such as one-offpayments for the birth of third and successive children, as well as for multiple birth, were alsoimplemented. Their impact in the non-contributory benefits awarded by the Social Security is verylimited.

Regional Minimum Income Schemes (Rentas Mínimas de Inserción)

These are non-contributory programmes regionally implemented by the seventeen Spanish ComunidadesAutónomas with different characteristics but with the same general aim of social insertion for low-incomeor non-income families. Amounts of minimum income guaranteed vary according to the “generosity” ofeach region. Access to these programmes and benefits is available for families whose income is lower thanestablished thresholds in each region (in some of them there is an annual budgetary limit for the totalamount of benefits to be awarded). Programme eligibility and continuity for beneficiaries are conditional ontheir engagement in activities of social insertion. According to data collected by the Spanish GeneralDirectorate of Social Action, Minors and Family (Dirección General de Acción Social, del Menor y de laFamilia), 78,645 families were receiving monthly these benefits in 2000 (for a more detailed analysis, see p.000).

A fragmented “safety net”

Despite its fragmentation, an organizing rationale can be identified in the configuration of the public safetynet protection in Spain. Figure 4.2 illustrates the various levels and means of this basic social protection.

Note that all benefits for the low-income are means-tested, although how this criterion is applied varies insome aspects. For instance, some benefits are paid to people already receiving contributory pensions (SocialSecurity Minimum Pension Supplements), or to unemployed who have exhausted their contributoryunemployment benefit period (Social Assistance Benefits for the Unemployed). Benefits in cases of theNon-contributory Social Security Pensions, Family Benefits of the Social Security and Regional MinimumIncome subsidies can be claimed with no previous contribution to the Social Security system. Most of thesecash benefits take into account the aggregate level of all-family income (or household) as a criterion foreligibility. Further to this, some subsidies are differential benefits that increase existing income to anestablished minimum, whereas others are provided as final amounts.

All these benefits provide cash amount which are lower than the legally established minimum wage (seeTable 4.3). In other words, they provide a lower level of protection than that beneficiaries could get werethey to be active and employed in the formal labour market.34 The different benefits are ordered rangingfrom those which offer a better coverage to those more limited (in duration and intensity), and moreconditioned with the compliance of activities related generally to labour activation. Variations in theawarding of the benefits depend on institutional milieux and civil servants’ discretionary criteria (at all threestate, regional and local levels of government). Somehow the intensity of the benefits is inversely coherentwith the principle of territorial subsidiarity: the “better quality” benefits are also the ones provided by thecentral institutions (i.e. those of the contributory social security system). It is also noticeable that access to

SPAIN: POVERTY AND “SAFETY NETS” 135

“better quality” benefits is independent of the family income,35 while the lower levels are family benefitsstrictu sensu.

The contributory principle is the main criterion around which all different sub-systems are organized and,thus, a relationship with the formal labour market background of the beneficiaries is established. All thoseapplicants to low-income benefits who have secured a sufficient contribution to the Social Security can havebetter contributory benefits (retirement pensions, disability and survivors’ benefits, as well asunemployment benefits). When the contributory record of the claimant has come to a stop (unemployment),or is insufficient (pensions), means-tested benefits are to be supplemented so that a legally establishedminimum can be reached.

In the event of no previous contribution to the Social Security, a preferential criterion is set alsoaccording to the labour market situation of the claimant (age or degree of invalidity) as happens withcontributory benefits. For those workers who are not potentially active (over 65 years and disability over 65per cent), Social Security Non-Contributory pensions as well as LISMI and Assistance benefits are alsoavailable. Those claimants potentially active (under 65 years and with disability lower than 65 per cent) canalso be eligible under the regional programmes of minimum income.

Age is an important protection marker for the grouping of claimants in different categories. The age of 65years delimits active and non-active status. However, unemployment subsidies also take into account thespecific situation of those claimants over 45 and 52 years of age in order to allow for a relaxation of theeligibility criteria, the extension of the benefit duration, and other related circumstances. At the other end,for needy claimants under the age of 25, benefits have the family as unit of reference (exception made forthose already in the formal labour market or who have their own family). Note, however, that familybenefits for dependent children have as upper limit the age of 18 years. (Table 4.4 reproduces a list of mainfeatures of national legislation on benefits for the low-income.)

As regards the regional minimum income schemes, it has to be pointed out their residual or subsidiaryrole with respect to schemes of insertion mainly in the Social Security system. They have taken on board

Figure 4.2 The Spanish public “safety net”: levels and means of protection.

136 ANA ARRIBA AND LUIS MORENO

beneficiaries sent to them by programme reduction, changes and insufficiencies in the main socialassistance protection schemes (contributory and noncontributory). This regards not only incomemaintenance but also other policies on education, housing or health (Serrano and Arriba 1998).

The Spanish NAPSIs: Plan Nacional de Acción para la Inclusión Social del Reino deEspaña

Following the commitment expressed at the Lisbon European Council (June 2000) and the EU objectivesagreed at the European summits of Nice and Copenhagen, the Spanish Government engaged in the workingout of the National Plans on Social Inclusion (NAPSIs) for the periods 2001–03 and 2003–05. Co-ordination for the working out of both Plans has been the responsibility of a Secretariat organically locatedat the General Sub-Directorate of Programmes of Social Services, within the General Directorate for SocialAction, Minors and Family, in the Spanish Ministry of Labour and Social Affairs.36

As a result of the fragmented nature of the system of social protection in Spain, and the deep process ofdecentralization of welfare competencies, the central layer of government can be regarded as the oneplaying a subsidiary role with respect to the regions in matters of social welfare. The Secretariat did nothave an easy task in securing information provided horizontally by other ministerial bodies of the centralgovernmental, or by others layers of government (regional, principally), with regard to other sector policies(education, employment or housing). It was not the task of the Secretariat to make fresh inputs into thePlans, and its human and material resources were limited.37

As expressed by the officials involved,38 the main task for the elaboration of the Plans was basedprimarily on co-ordination activities. Accordingly, the Plan Secretariat counted on the participation andcontribution provided by other institutions actively involved in the fight against poverty and exclusion: (i)public bodies (central, regional and local governments); (ii) social partners (trade unions and Economic andSocial Council);39 (iii) NGOs,40 (iv) experts41 and (v) targeted groups and “users”.

For the elaboration of the 2001–03 Plan, information on the activities and proposals related to the fourobjectives established in EU’s Nice Summit was compiled. In the 2003–05 Plan, data for the evaluation ofthe previous Plan were also included. For each planned action, characteristics of the less-favoured groups,number of beneficiaries, human and budgetary resources were requested. The collected information wasaggregated and systematized into what are the central contents of the NAPSI and the evaluation of 2001–03.

Thus, the Spanish 2001–03 NAPSI did not incorporate new strategies or a reorientation of actions alreadyestablished by the various governmental bodies involved in the fight against social exclusion. It can beregarded as the very first attempt made by one central government institution to coordinate information atboth horizontal and vertical levels in social matters. In this respect, the most relevant achievement of thefirst Spanish NAPSI was active intergovernmental participation and the involvement of social partners andstakeholders. Different perspectives were expressed by the participants, which indicated the difficulties inarticulating common institutional grounds in the fight against poverty and social exclusion in Spain. As aresult, an exhaustive recompilation of the various policies and programmes implemented at the variouslayers and institutions of government and by the concerned civil society organizations was produced. Thesame pattern was basically followed with the 2003–05 Plan.

The Spanish NAPSIs did not put forward new ideas. They provide a picture of what is already being doneby the different social actors and stakeholders involved in programmes of social inclusion. The existingstructures of social protection were not scrutinized or even questioned in the NAPSIs. Consequently, a gooddeal of actions, measures and proposals included in both Plans make references to already existing

SPAIN: POVERTY AND “SAFETY NETS” 137

governmental plans, e.g. Employment Plan (Plan Nacional de Empleo);42 the Concerted Plan for theDevelopment of the Basic Provision of Social Services by the Local Authorities (Plan Concertado para elDesarrollo de Prestaciones Básicas de Servicios Sociales de las Corporaciones Locales); the Integral Planfor Family Support (Plan Integral de Apoyo a la Familia); the Equal Opportunities Plan for Women (Plande Igualdad de Oportunidades); the Integral Plan Against Domestic Violence (Plan Integral contra laViolencia Doméstica); the Professional Insertion Plans (Planes de Inserción Profesional); the National Planof Housing (Plan Nacional de Vivienda); access to new technologies (Plan Info XXI); the Action Plan forElderly People (Plan de Acción para las Personas Mayores); the Action Plan for the Disabled (Plan deAcción para Personas con Discapacidad); or reinforcement of the pensions system, according to theagreement on improvement of social protection reached by the employers’ associations and the trade unionCCOO on April, 2001.

As a result, the Spanish NAPSIs lacked a clear delimitation concerning the comprehensive domain ofsocial exclusion (in some cases related to processes of societal change, in others to problems of social cohesionsuch as those derived from unemployment, or regarding groups with problems of marginalization). Therewas not a unitary definition of a set of priorities which could facilitate social inclusion. Rather, the NAPSIsoffer a mosaic of heterogeneous measures and proposals which in most cases reflect the proposals putforward by the social actors to combat social exclusion.43

None of the Spanish NAPSIs’ actions counted on extra financing to accomplish its stated aims. The mainreason for this austere rationale has to be found in the general budgetary aim embraced by most PPGovernment policies: the “zero deficit”. Consequently, no extra funding or budgetary provision was to beincorporated in the Plan. Public financing of actions was already included in the corresponding budgetaryentries of the bodies involved (e.g. NGOs’ programmes continued to be carried out according to theavailable money collected by income tax). A look at the financial data for the 2001–03 and 2003–05periods, reproduced in Table 4.5, shows that the larger budgetary amounts or increases correspond toprogrammes already established and developed as priorities by the central

Table 4.5 National Action Plans’ aggregate budgetary efforts (2001–04) (� million)

Objective/Area of action 2000 accomplished (revised) 2001+2002 accomplished(provisional)

2003+2004 estimated

Objective 1.1 1,071.2 2,531.1 2,933.6

Access to employment 1,071.2 2,531.1 2,933.6

Objective 1.2 13,226.8 28,152.0 29,332.1

Income guarantee 10,107.2 20,534.3 21,125.8

Social Services 534.6 1,171.6 1,243.5

Access to housing 446.8 1,030.4 1,113.0

Education 1,064.7 3,005.1 3,227.2

Health 1,073.5 2,410.7 2,565.1

Justice – – 57.5

Objective 2 327.6 798.0 702.1

Inclusion/territorialinterventions

– – 14.1

Support family solidarity 315.0 728.9 616.3

Access to new technologies 12.6 69.1 71.7

Objective 3 849.1 2,076.4 2,478.8

138 ANA ARRIBA AND LUIS MORENO

Objective/Area of action 2000 accomplished (revised) 2001+2002 accomplished(provisional)

2003+2004 estimated

Elderly 176.3 636.1 930.2

Disabled 235.6 517.6 564.9

Women 62.0 134.1 140.8

Youth 42.5 95.2 106.9

Children and family 197.7 400.4 403.6

Gypsy population 60.1 120.7 126.1

Immigrants 63.2 149.3 182.2

Homeless 8.6 17.3 18.2

Former inmates 3.1 5.8 6.0

Objective 4 102.6 259.7 288.4

Mobilization of agents 102.6 259.7 288.4

Total 15,577.2 33,817.1 35,735.0

Sources: Spanish Ministry of Labour and Social Affairs (MTAS 2003:39 and Appendix II, 94).

government. These programmes were mainly in either social policy matters (employment, family support orelderly care), or in new fields of intervention (new technologies), which were now “incorporated” in theNAPSI as actions favouring social inclusion (Rodríguez-Cabrero et al. 2003).

The 2003–05 Plan largely follows the methodology and actions initiated by the first NAPSI.44 The latteris not to be regarded as a platform for a multilateral impulse facilitating social inclusion, but it has beeneffective, according to its evaluators, in expanding the concept of social exclusion among the various actorsundertaking social policy development (MTAS 2003).

The methodology for evaluating the 2001–03 Plan was similar to its initial elaboration: the final reportwas the result of an aggregation of the various assessment inputs produced by the intervening evaluators. Ingeneral terms, the appraisal is very positive, which is reflected in the document with an increase ofactivities, targeted groups and users, as well as an optimization of expenditure. However, some notableshortcomings in this evaluation exercise can be pointed out:

• NAPSIs’ assessment is based on the enumeration of actions and measures taken without analyses of thereal effects. If, for example, there is data on the number of people who had access to the Plan regardingmeasures to improve their employability, no figures are provided on how many individuals finally founda job. In this respect, the evaluation is rather on Plan’s output than on outcomes.45

• There is no identification of the weak points and main obstacles for the actual implementation of the originalPlan. Likewise, there is a lack of self-criticism in the actual shortcomings faced by the running of thePlan.

• There is no comparison between actions originally planned and actually implemented. Furthermore,measures which were not stated initially were evaluated whereas others included originally were notsubject to appraisal.

• Impact assessment of actions and measures was difficult in many cases because they were incorporatedfrom existing programmes which had been put into effect without regarding the NAPSI. There is littlereason to claim, for instance, that improvement in child education was due to the NAPSI.

SPAIN: POVERTY AND “SAFETY NETS” 139

• There is a lack of evaluating criteria and priorities. The increase in the number of individuals who hadaccess to goods and resources through “non-normalized” circuits (Programas de Garantía Social) is notper se a good result unless it is initially stated.

• There seems to be a clear interference between planning and evaluating. Members of the same team forplanning carried out evaluation activities as well.

Let us briefly analyse what were considered to be the main contributions and innovations of the SpanishNAPSI 2001–03, and which continue to appear as crucial factors in the 2003–05 Plan. First, the elaborationof the Territorial Action Plans for Social Inclusion (or TAPSIs), a commitment assumed by the seventeenSpanish Comunidades Autónomas,46 with a particular reference to Local Plans for Social Inclusion to bedeveloped by municipalities representing up to 40 per cent of the total Spanish population. In some of them,the TAPSIs had already been elaborated (Basque Country, Canary Islands, Castille and Leon, Castille-LaMancha, Catalonia, La Rioja and Navarre) .47 Others were in the process of development, such as that ofMadrid. According to the evaluation of the Plan, in September 2003 thirteen TAPSIs had already beenapproved, although they were not listed in the document. There was no explanation why the remainingregions had not been able to finalize theirs. Information on TAPSIs’ main orientation, characteristics andfeatures was also lacking.

The second contribution referred to the crucial importance that mobilization by social actors andstakeholders should have for the future viability of the Spanish NAPSI. The Plan foresaw the establishmentof several multi-level committees and commissions for the optimization of information exchange and groupparticipation, including NGOs, experts, practitioners and media leaders. Participation of civil societyorganizations involved representatives of the Red Cross, Caritas, the Gypsy Secretariat and the CERMI(Comité Español de Representantes de Minusválidos, Spanish Committee of Disabled Representatives).These NGOs are equipped with better material infrastructure and human resources than other “low-profile”associations. Accordingly, they were able to make not only comprehensive proposals, but could alsoarticulate strong critical views, something which other rank-and-file associations could not afford.48

The document of evaluation of the NAPSI positively assessed the monitoring of activities throughcommittees, co-ordinating bodies or debating forums. However, little was said on the outcomes of suchactivities or the obstacles and limitations faced in the participation process.49

The 2003–04 Plans insist on the importance of mobilizing not only social actors and agents, but alsotargeting groups and users.50 In fact, this emphasis on participation is reflected in the incorporation ofassociations which incorporate targeted groups in the evaluation process.51 Institutions of the Third Sectorappear in such a manner as social “double agents” with intervening roles tout court first, they are managersand promoters of programmes for social inclusion (or NGOs in the NAPSI’s terminology); and second, theyare mediators—or, rather, intermediaries—between the public institutions and the potential “users” in risksituations (they consider themselves as associations of targeted groups according to the denomination of theNAPSI).

In sum, resource co-ordination and data systematization in the fight against social exclusion areparamount concerns in a quasi-federal country such as Spain. The process of development and evaluation ofthe Spanish NAPSIs has provided a great opportunity to advance along such a route. However, the first setof results seem to indicate that greater involvement of the various layers of government and the social actorsand agents involved in facilitating social inclusion is required. The aggregation of activities to enhancesocial inclusion is, therefore, needed in order to achieve common objectives.

140 ANA ARRIBA AND LUIS MORENO

Regional minimum incomes schemes

The Spanish mesogovernments of the Comunidades Autónomas have shown an active interest in policyinnovation concerning welfare programmes. Among the various actions taken by them one is to beidentified as having far-reaching repercussions for the “safety nets” of social protection in Spain: theprogrammes of minimum income of insertion (Rentas Mínimas de Inserción).

The general process started in September of 1988 with the announcement by the Basque Government of aregional Plan de Lucha contra la Pobreza (Programme against Poverty). This initiative sparked off a regionalmimesis, or “demonstration effect”, on the part of the other sixteen Comunidades Autónomas. By the end of1990s, all Spanish mesogovernments were engaged in the implementation of regional programmes ofminimum income.

Some of the programmes of minimum income were established mainly on the initiative of themesogovernments, and as a result of the combined action by both types of policy-makers (electedpoliticians and executive officials). In other cases, the pressure exerted by the opposition parties in theregional parliaments was the main factor behind the development of these programmes. Finally, a third pathof policy-making was due to the mobilization of the regional branches of the main trade unions (CCOO andUGT), as well as some significant NGOs, such as Caritas.

As a result of this process, since the early 1990s all Comunidades Autónomas have undertaken socialassistance programmes for guaranteed minimum income. Due to the fact that they were all included in theregional political agendas and that the common rhetoric for the implementation of these programmes was infavour of social inclusion, we will analyse them as a common block of policies. Certainly, there is a degreeof internal diversity within this block, something which adds just another element of fragmentation in theSpanish safety net of social protection. Some Comunidades Autónomas have been able to develop benefitsof social minima which can be considered as genuine income guaranteed by regional laws. Others just offerassistance benefits of a low intensity and coverage, while others are more discretional. All thingsconsidered, and in contrast with Italy, where the survival of the programme of reddito minimo d’inserimentowas mainly dependent on resources granted by the central government, the Rentas Mínimas de Insercióncan be regarded as an autonomous initiative taken by the Spanish mesogovernments with a commonpolitical discourse although with different policy outputs.

Main features of the regional programmes

As a proviso, the lack of comprehensive and systematic information compiled on these programmes shouldbe acknowledged. There is a temporal mismatch on the available information which makes systematicanalyses difficult. Regional sources are many and do not always follow similar patterns of data collection.Thus, analyses will mainly take into account assessments mainly made on the first period of implementationof regional Rentas Mínimas de Inserción (1989–94), when research was carried out based upon primarydata collected with consistent criteria (Aguilar et al. 1995; Ayala 2000). We will make use of data laterproduced systematically by the General Directorate of Social Action, Childhood and Family (DirecciónGeneral de Acción Social, del Menor y la Familia) but on the information facilitated by the ComunidadesAutónomas, not always collected with similar indicators.52 The latter should therefore be taken with adegree of caution.

The regional programmes of minimum income have distinct characteristics with regard to the intensity ofthe benefits and the insertion obligations to be followed by the beneficiaries. However, they are quasi-universalistic entitlements, which combine cash benefits with activation policies and programmes of social

SPAIN: POVERTY AND “SAFETY NETS” 141

integration (employment promotion and vocational training courses, primarily). Their main commonfeatures can be identified as follows:

1 Families are the units of reference even though individuals can be single beneficiaries.2 Means-tested criteria are related to a threshold of household income below which cash benefits are

awarded (around two-thirds of the minimum wage).3 Residence status of applicants is required (ranging from one to ten years).4 Periods of extension are available provided that beneficiaries have complied with the social insertion

activities and the social needs remain the same.

Indeed, policy outcomes have resulted in some visible differences in the intensity of the benefits and, aboveall, the nature of the “insertion” programmes to be accomplished by the beneficiaries. According to thesedifferences, three groups of the regional programme of Rentas Mínimas de Inserción can be identified(Laparra and Aguilar 1997):

1 those establishing a link between the perception of the subsidy and the aim of insertion; 2 those where insertion takes the form of a job contract, and the benefits are subject to workfare;3 those providing discretionary and ad hoc social assistance.

Together with this criterion, the levels of coverage and the protecting intensity of the programme benefits53

have been taken into account in order to establish the typology shown in Table 4.6. The informationreproduced in Table 4.6 relates to three large groups. The programme implemented in the Basque Countryis the one which can be considered a “genuine” minimum income scheme of insertion. As reproduced abovethese lines, there are other programmes which could be regarded as minimum incomes schemes with somelimitations, due to restrictive elements, mostly of a budgetary nature. The second group refers to minimumsubsidies with legal restrictions (requirements for programme access and benefit continuation), and a finalone is characterized by a very limited coverage and intensity. Further to this, there are workfareprogrammes which prioritize temporary jobs of “social usefulness”, and which residually provide periodicalcash benefits. In any case, this classification should be re-assessed as several of the regional programmeshave been adjusted and modified in recent years. At present an evaluative exercise taking into account allregional programmes is very much needed.

Often it is asked whether or not the RMIs programmes imply a citizen’s right. The promulgation of aregulatory regional law could be adduced for the claiming of a legal entitlement to be contested before thecourt. In

Table 4.6 Typology of the regional programmes of minimum income of insertion

Coverage and protectingintensity

Benefit and insertionprogramme

Social assistanceprogrammes

Protected socialemployment

High Basque Country

Medium Madrid Aragona La-La Mancha

Catalonia Galicia Asturias

Navarreb Murciaa La Riojaa

Low Balearicsb Andalusia

Canaries

142 ANA ARRIBA AND LUIS MORENO

Coverage and protectingintensity

Benefit and insertionprogramme

Social assistanceprogrammes

Protected socialemployment

Cantabria

Castile and Leon

Valencia

Extremadura

Sources: Based on Aguilar et al. (1995).Notesa The small amount of the benefits involved is to some extent compensated by other subsidies of family integration.b Based on information provided after 1995. Underlined: Established by regional law (2002).

Table 4.6 it has underlined those regions which had passed regional acts of such a nature up until the year2000 (to which the 2001 law passed in Madrid Region should be added). However, and despite thelegislative status, the individual legal entitlement has its limits on the budgetary availability of funds tocover the due expenses to be claimed by the potential beneficiary (Table 4.7). The exception to this generalconsideration is the Basque programme, which provides full entitlements disregarding budgetary limitations.

In explaining the differences in policy outcomes, the variable financial manoeuvrability has been adducedas the main explanatory factor. Certainly, the Basque Country and Navarre with a system of fiscal quasi-independence have been able to fund more generously their programmes of minimum income. Note that ascompared with the autonomous public spending in Catalonia, the Basque per capita expenditure is 1.8 higher.54

Nevertheless, the setting of political priorities in policy funding appears to be the most compellingexplanatory element. After all, the Spanish mesogovernments have the final budgetary say in the running ofthe kind of programmes which are the product of their own initiative.

Access to programmes and criteria of eligibility

In the majority of the Comunidades Autónomas an annual amount is fixed to cover the expenditureassociated with the programme and, once this is exhausted, no more beneficiaries can be accepted. In someother regions, where the entitlement to the minimum income is guaranteed, a set of indicators is establishedfor claimants. In this section, we refer to accessibility and eligibility criteria common to most of theregional programmes. Specific requisites to comply with are as follows:

• Nationality and residence. Nationality is not a requirement for access to minimum income programmesin most regions (with the exception of Cantabria and the Canary Islands). However, a certain period ofregistered residence by the beneficiary in the Comunidad Autónoma is required in all programmes.Initially, regional policy-makers feared of some kind of “welfare tourism” (beneficiaries moving aroundthe country to apply for the most generous subsidies), although this has not proved to be the case. Insome regions the residence precondition is a clear disincentive for the illegal immigration in a policydomain in which the central administration has adopted a tough “law-and-order” approach, and there is aclear absence of a comprehensive intergovernmental plan of action. Requirements vary from theobligation for the potential beneficiary to have resided since a concrete date (e.g. 1985 in Cantabria), tolong periods of residence (e.g. ten years in Navarre or five years in Galicia). In some Comunidades theprerequisite of residence is shorter and ranges from two or three years in

SPAIN: POVERTY AND “SAFETY NETS” 143

Table 4.7 Regional expenditure and coverage of the programmes of minimum income schemes (2000)

Autonomouscommunities

Regional expenditure Population coverage Protective intensity

Actualspending(million of� )a

(%)Budgetary effortb

Numberofhouseholdsc

(%) ofhouseholdsd

BasicAmounte (� /month)

(%)M.W.f

(%)N.C.P.g

Andalusia

27.5 0.17 15,962 0.73 263.4 53.1 91.4

Aragon 2.6 0.12 1,396 0.33 254.8 51.4 88.5

Asturias 10.6 0.86 1,512 0.45 282.3 57.0 98.0

Balearics 1.1 0.12 570 0.21 282.1 56.9 97.9

BasqueCountry

53.2 1.03 16,550 0.52 305.4 61.6 106.0

Canaries 8.0 0.21 5,358 1.12 238.9 48.2 82.9

Cantabria 3.5 0.39 2,340 1.39 249.4 50.3 86.6

Castile-LaMancha

1.9 0.07 813 0.15 297.5 60.0 103.3

Castileand Leon

8.8 0.20 2,814 0.32 260.3 52.5 90.3

Catalonia 36.6 0.28 9,726 0.47 285.5 57.6 99.1

Extremadura

3.1 0.18 1,351 0.38 318.6 64.3 110.6

Galicia 12.7 0.20 4,156 0.50 242.0 48.8 84.0

La Rioja 0.3 0.08 179 0.22 289.1 58.3 100.3

Madrid 24.9 0.38 7,855 0.48 249.0 50.2 86.4

Murcia 1.0 0.06 3,478 1.00 240.4 48.5 83.4

Navarre 4.1 0.19 1,820 1.02 318.6 64.3 110.6

Valencia 10.0 0.14 2,565 0.20 298.0 60.1 103.4

Total 210.0 – 78,445 0.61 – – –

Averages – 0.28 – – 275.0 55.5 95.5

Sources: Elaboration from data provided by the General Directorate for Social Action, Minors and Family(Direction General de Acción Social, del Menor y de la Familia) (Spanish Ministry of Labour and SocialAffairs).

Notesa Budget expenditure amounts correspond to minimum income payments, except for Asturias which also

includes social wage expenditure and training,b Percentage of expenditure related to minimum income schemes as percentage of the consolidated regional

budgets of the Comunidades Autónomas, Ministry of Public Administration (http://www.map.es)c The number of households correspond to the number of benefits. In some cases, data correspond to the number

of households with access to the programme at any moment (e.g. during December in Madrid); in

144 ANA ARRIBA AND LUIS MORENO

some others, data correspond to the number of households with access to the benefits throughout theyear (e.g. Basque Country),

d Source for the total number of households of the Autonomous Communities is the Survey on Family Accounts(ECPF), 4th Trimester, 2000, INE (Spanish National Institute of Statistics).

e Basic initial amounts. The final benefit to be perceived by the beneficiaries is the result of adding familysupplements (depending on family size and on each CCAA) and subtracting any income received bythe family,

f Percentage of the basic amount of the minimum income with regard to the legally established minimum wage(year 2000:495.5 � /month; annual amount apportioned in 12 payments),

g Percentage of the basic amount of the minimum income with regard to non-contributory pension for disabilityand old-age (year 2000:288.10 for 12 months; annual amount apportioned in 14 payments).

Catalonia and La Rioja, to one year in Valencia, or none in the case of Castille-La Mancha.• Household formation and composition. Regional minimum income benefits have families as units of

reference. The application is submitted by an individual who, once the benefit is awarded, is responsiblefor its family distribution. In consequence, all members of the family have access to the subsidy throughone direct claimant/beneficiary.55 There are several definitions of what a “family unit” is and diversedegrees of lineage are taken into account. The diverse effects of these family definitions are difficult tomeasure, but two are worth mentioning: (i) definitions of a more extended type of family imply thepossible inclusion of other relatives’ income; and (ii) the possibility of receiving two or more subsidies inone single household.56

In all regional schemes there is no discrimination concerning single households, although in someComunidades this type of household is considered an “exception” rather than a rule. As a matter of fact,single household beneficiaries form one of the largest groups of beneficiaries of the regional minimumincome schemes.

Regional norms also make provision concerning family formation. Requirements in this respect aim atavoiding “artificial” family creation in order to qualify for minimum income benefits. Prerequisites rangefrom six to twelve months for an “independent” household to be eligible. In some regions such arequirement is exempted, or in the case that households are the result of family ruptures.

• Age. Most regional requirements set age limits between 25 and 65 years. Protection for citizens over 65years rests mainly with noncontributory pensions. In this way, regional minimum income plays asubsidiary role with respect to Social Security benefits.57 On adopting the lower age limit fixed by theFrench RMI (25 years),58 it is implicitly acknowledged the supporting role of the family until the youngfamily member is expected to be in a position of economic emancipation, i.e. he/she has entered the formallabour market and is in a position to establish his/her own family (Aguilar et al. 1995). Implicit in thesetting of this age is a certain attitude of “paternalistic” reluctance towards the kind of uses thatyoungsters under 25 years could put these cash benefits.

• Economic resources of the beneficiaries. In all Comunidades Autónomas access to the benefits is clearedonce it is tested that family income is lower to the amount of the renta mínima to be perceived (seeTable 4.7). In some Comunidades Autónomas, the basic amount is established automatically in referenceto the minimum wage. In others, the amount of reference is that of the mean non-contributory pension.However, in most cases the amount is fixed in the annual budget by the regional parliament.

In those programmes associated with “protected special employ ment” (see Table 4.6), various optionsare arranged to calculate the level of the subsidy (higher, in the case of Navarre, and lower in Andalusia,Asturias, and Castille-La Mancha).

Basic amount supplements are added for each of the family members (according to equivalencescales). In most regional programmes supplements are calculated according to the number of family

SPAIN: POVERTY AND “SAFETY NETS” 145

members with disregard to their composition (e.g. adults, minors or disabled). Supplement amountsdecrease as the number of family members increase. In some cases maximum limits have beenestablished according to two modalities: (i) once a number of family members is reached, there is nofurther supplement; (ii) an aggregate top amount is put as a limit. Let us remember that, in general,equivalence ratios are very flat (usually the amount for the second family member does not go above 30per cent of the head of the family). Note that from the final calculated amount of renta mínima, anyavailable income of the family is deducted.

• Means-testing. Programmes of rentas mínimas have different methods to take into account familyincome, particularly that from labour and other social benefits in order to avoid a reduction of protection,or to produce labour disincentives. Accordingly, labour income is subtracted partially in some regionalprogrammes, whereas in others an exempted amount is established (Aragon). Concerning other low-income benefits, most Comunidades Autónomas subtract integrally the amount perceived as non-contributory pension by the beneficiary. Some others do it partially as concerns income from socialsubsidies perceived by other family members, and in order to prevent unwanted effects of family break-up (e.g. Madrid). Concerning contributory family benefits some regional programmes consider themincompatible with the renta mínima, while others subtract them or, conversely, do not take them intoaccount (Madrid, Catalonia). As regards those benefits targeted to cover specific needs (e.g.scholarships, transport allowances, medicine payments, emergency help, etc.), the common procedure isnot to take them into account in the calculation of the amount of the minimum income benefit.

Most programmes also refer to patrimonial resources. In some instances they are assessed according toits gains (e.g. Madrid or Asturias). In others, they are used as indicators of sufficient means of standardof living and, in general, are considered to be incompatible with the awarding of the renta mínima. Inmost cases the value of owned houses where beneficiaries live is not taken into account.

While in some Comunidades Autónomas the general criteria is to discard a strict regime ofincompatibilities, in others these are established in a more contingent manner, and taking into accountthe type of labour activities of the claimant, the family and social benefits already perceived, and so forth.

• Commitments and obligations. All regional programmes establish a number of obligations to beaccomplished by the beneficiaries. Basically they refer to the commitment to use the renta mínima tocover basic needs of the household, to report any variations in the family unit, to reimburse any undueincome and to apply to other subsidies to which the beneficiary may be entitled.

The aim of insertion embraced by all regional programmes is articulated in some programmes as a “doubleentitlement” and in others as a “counter-benefit” (contraprestación), or reciprocity by the beneficiarytowards the community. In some schemes commitments are requested not to refuse to take up a job offer orto carry out activities, either workfare or “social”. Other requirements include the prohibition of publicbegging, the compulsory schooling of children, the registration in job centres and the adoption of a“favourable” attitude towards insertion.

Programme provision and administration

Matters concerning programme management are also heterogeneous. Claimants need first to make anapplication in order for the process to be initiated. Usually the application is submitted at the municipalunits of community care (unidades de atención primaria de servicios sociales), as well as at the regionaloffices of social services. In Catalonia some non-profit making NGOs are entitled to manage theprogramme. On applying, claimants must produce a number of documents to support the compliance of

146 ANA ARRIBA AND LUIS MORENO

programme requirements. In some cases this first step appears to be insurmountable because of theproblems some claimants have in fully understanding the requested information (Serrano and Arriba 1998).

The decision process on the applications is final after a fixed period of assessment (something which notalways is accomplished). From the date of submission until the actual receipt of the first payment theduration of the process can vary from four to eight months (Ayala 2001). Programme monitoring andsurveillance of the obligations to be accomplished by the beneficiaries are responsibilities of the local socialservices.

In some Comunidades Autónomas the continuation of the programme is simply conditional on themaintenance of the initial circumstances of the beneficiaries (e.g. the Basque Country). In others, there is amaximum period of six months within a fiscal year. In most cases, however, the award of the benefit is for sixmonths which can be extended to a similar period, although some regional programmes do not includeautomatic extension.

Programme managers face no small problem with means-testing. There are not systematic data on thefunctioning of this access mechanism, nor on the “unwanted” effects which have usually been associatedwith its implementation (e.g. difficulties in proving the income status of the potential beneficiary, or thelevel of fraud and false declarations). Based on information provided by research of a qualitative nature, itcan be speculated that the lack of data concerning this mechanism is due to the unease of the programmemanagers to provide an evaluation which could jeopardize what they believe it is still a low level of popularlegitimacy. Participant observation in the running of these RMIs programmes has corroborated that a degreeof discretionary functioning allows the programme managers to “adapt” the normative requirements to theneeds and personal characteristics of the applicants and potential beneficiaries (Serrano and Arriba 1998).

Insertion activities

In the mid-1990s59 only the Basque programme was considered a “genuine” programme simultaneouslyguaranteeing the “double entitlement” to cash benefits and insertion. In the rest of the cases both cashbenefits and insertion activities appear as differentiated elements. In general, the latter is subordinated to theformer in various degrees and manners. This relationship of subordination materializes with the signing of acontract by the beneficiaries.

Insertion activities are usually individualized. In many Comunidades Autónomas they are determined atdifferent administrative levels of the regional systems of social services. However, in some instancesparticipation of beneficiaries in working out the kind of individualized activities of insertion is required.

In most cases social insertion is related to the carrying-out of activities of a productive nature.Classification of activities can be distinguished as follows (Aguilar et al. 1995):

• Area of social services and social work: personal promotion, access to systems of social protection,eradication of begging, and access to occupational, leisure and/or therapeutical activities.

• Area of education: further adult education and instruction for minors.• Area of employment: orientation and motivation, professional training, workfare incentives, self-

employment and social economy, social employment and insertion companies.• Area of co-operation: community help and care for the elderly.

A good deal of these insertion activities falls into other department competencies (education, employment,health-or housing). Some regional programmes do not take into account—at least explicitly—anyinterdepartmental co-ordination or joint committee for this purpose. In other cases, programme administrators

SPAIN: POVERTY AND “SAFETY NETS” 147

have found problems of horizontal coordination along the lines of what has been previously commentedregarding the development of the Spanish NAPSI.

Minimum income programmes include mechanisms for specific insertion needs. Among them we refer tothe so-called “projects of insertion”. They are programmes with different labels and designs which sharesome common features: (i) they generally seek to finance targeted non-profit organizations; (ii) activitiesare targeted for excluded groups, usually beneficiaries of minimum income programmes; and (iii) insertionactivities mainly include professional training and orientation to enter the labour market. Their scope islimited although their main input has been to provide a space for collaboration with non-profit associations,to carry out intensive work with some groups that are especially vulnerable and to allow for theexperimentation of new forms of support for social insertion. Among them new companies of insertion haveemerged which operate in the “open market” and have employed workers with a previous record ofexclusion.

Policy implementation

There have been different actors shaping policy inputs and design of the programmes of Rentas Mínimas deInserción. However, the mesogovernments of the Spanish Comunidades Autónomas are to be regarded as themain protagonists in their implementation. Let us not forget that, prior to approval by the regionalparliaments, no explicit popular demand was expressed in any of the Comunidades Autónomas to encouragetheir implementation. The institutional factor that made relevant the issue of the guaranteed minimumincome was precisely the constitutional entitlement for the regions to exercise their political autonomy. Noreactive “path dependency” could be referred to in this case. This fact validates the assumption that policyinnovation concerning social policies developed by sub-state communities with a “cosmopolitan localism”perspective can be more effective and efficient (Moreno 2003).

The central Ministry for Social Affairs showed no small reticence in the implementation of the regionalprogrammes of minimum income. Its main reluctance concerned arguments of poverty dependency andlabour disincentives. Allegations that these new regional policies could affect territorial solidaritythroughout Spain were among its criticisms (Ministerio de Asuntos Sociales 1989). However, the thennewly-created Ministry had already opted to make the generalization of the non-contributory pensions itsmain priority, exhausting in this ambitious programme most of its political capital within the centralgovernment. The initiative taken by the Comunidades Autónomas left little room for the institutionalmanoeuvring of the central Ministry, and was based on those constitutional provisions safeguardingregional self-government. Within this context, no action to boycott the implementation of the regionalprogrammes of minimum income was to be expected from the PSOE central government.

Since the beginning of the process, different arguments in favour or against the implementation of theregional Rentas Mínimas de Inserción have been neither clear nor sophisticated in their analyses (Aliena1991). In this respect, the “simplicity” of the trend-setting arguments used by the Basque policy-makers isrepresentative:

1 The Rentas Mínimas de Inserción were to overcome social marginalization.2 No labour passivity was to be encouraged.3 EU recommendations and other European experiences, such as the French Revenu Minimum

d’Insertion, lent support to the programme.4 There was a high degree of inter-party consensus and support from various Basque civil institutions.

148 ANA ARRIBA AND LUIS MORENO

Probably, among the factors above mentioned, the reference to EU recommendations is of particularimportance. There was constant reference to the proposals made by the European Commission encouragingthe recasting of the European systems of social protection so that guaranteed income could facilitate socialand labour insertion of poor and excluded citizens. This became the main line of argument for theformulation of the regional Rentas Mínimas de Inserción. Table 4.8 shows the timetable for the developmentand evaluation of the minimum income schemes in the seventeen Autonomous Communities.

Subsequently, the main trade unions supported these programmes. They pursued a course of actionalternative from the traditional negotiation and eventual agreement at the national level. In fact, the climateof national confrontation between the central government and the trade unions induced these negotiations atthe regional level. Note that, between autumn of 1989 and spring of 1990 all the pacts between themesogovernments and the trade unions (CCOO and UGT) included the establishment of programmes ofminimum income.

In parallel, a number of NGOs (Catholic Church institutions, principally) committed themselves tosupport the Rentas Mínimas de Inserción, both at regional level (Caritas parish councils), but also nation-wide (Spanish Caritas). They demanded the implementation of the programmes of minimum income but didnot participate in the forums where policy design was discussed (with the exception of Catalonia).

Information about the various modalities and characteristics of the regional programmes was circulatedamong the Comunidades Autónomas. Formal and ad hoc meetings took place in which policy-makers andexperts exchanged views and opinions on different aspects of the programmes to be implemented. Theirmain models of reference were the RMI, in France, and the “Programme against Poverty”, in the BasqueCountry.

The Comunidades Autónomas developed their programmes of rentas mínimas in various manners. In somecases, think-tanks of politicians, officials and experts within the organic structure of the regionaldepartments

Table 4.8 Minimum income schemes: main legislative dispositions

Autonomouscommunities

Development and implementation (D&I) Evaluation and reforms (E&R)

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Andalusia D&I E&R

Aragon D&I

Asturias D&I D&I

Balearics D&I

Basque Country D&I D&I E&R E&R

Canaries D&I E&R

Cantabria D&I E&R

La-La Mancha D&I D&I E&R

Castile and Leon D&I E&R

Catalonia D&I E&R

Extremadura D&I E&R

Galicia D&I E&R

La Rioja D&I

Madrid D&I E&R

SPAIN: POVERTY AND “SAFETY NETS” 149

Autonomouscommunities

Development and implementation (D&I) Evaluation and reforms (E&R)

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Murcia D&I

Navarre D&I E&R

Valencia D&I

of social policy prepared the pieces of legislation “behind closed doors”. In others, the process was open tothe inputs made by trade unions or NGOs. However, the ultimate decision on the development of theprogrammes remained with the mesogovernments.

In the phase of actual implementation of the programmes, the Comunidades Autónomas had to establishinstitutional agreements with the local authorities within their territories. The latter were to be the executivecornerstone in the programmes’ design. Such agreements, however, brought about some delays in theimplementation of the programme due to discrepancies stemming from the often different politicalcolouring of the local and regional administrations involved. This was evident in the case of large citiessuch as Barcelona controlled by the Left, but with a Centre-right nationalist coalition at the Generalitatgovernment, or Madrid, with a Right-to-centre City Hall and a Left regional administration. “Pretexts” forconflict and institutional warfare referred in most cases to problems of financing, as well as to the lack ofinfrastructure of social centres for primary assistance and community care.

In the process of policy implementation, other criticisms were put forward by the social workers andprogramme managers. Their initial attitude was one of general consent. Complaints about the excessivepaperwork and bureaucratic burdens were soon expressed. Likewise, the insufficient material infrastructurefor the management of the Rentas Mínimas de Inserción was also criticized. Moreover, the universalisticapproach of the social services during the 1980s was faced with the targeting criteria towards the needy putforward by the programmes of minimum income.

The process of policy implementation also confirmed the institutional leadership of the mesogovernmentsas main actors in the development of the Rentas Mínimas de Inserción. This factor was independent, inmany cases, of partisan alignments. Let us remember, for instance, that the Socialists were in a governmentcoalition with the Nationalists of the PNV in the Basque Country. They supported the first Renta Mínimaimplemented despite the reluctance of the Socialist Minister for Social Affairs. A similar situation tookplace in the region of Madrid. Thus, no consistent patterns of ideological standings are to be deduced fromthe political behaviour of the party organizations in the various institutional arenas of their participation.

The Comunidades Autónomas have followed patterns of mobilization rooted on a self-perceived“comparative grievance”: no region wants to be left behind. These perceptions have conflicted with the“differential fact” claimed by the Spanish “historical nationalities”: the Basque Country, Catalonia andGalicia. These regions are more interested in maintaining a “differentiated” degree of home rule ascompared with the rest of the Spanish Comunidades Autónomas. The combination of these processes hasresulted in a de facto policy equalization and in an incentive for policy innovation in those Spanish regionswhich have traditionally been lagging behind the “modernized” ones.

Assessment of impacts of minimum income schemes

Unsystematic data on regional minimum income constitute a problem when evaluating in depth the impactsof these programmes in each of the Comunidades Autónomas. However, in recent years the central General

150 ANA ARRIBA AND LUIS MORENO

Directorate for Social Action has been collecting series of basic data (benefits, amounts and budgets) fromwhich we can gain a picture of the general situation (see Tables 4.9 and 4.10).60

Information about basic amount for the calculation of minima benefits is a good indicator of theprotecting intensity of the various regional programmes. As reproduced in Table 4.7, basic amounts for theyear 2000 oscillated between a monthly payment of � 238.9 in the Canary Islands and � 318.6 inExtremadura and Navarre. This corresponded to 48.2 per cent and 64.3 per cent of the legal minimumwage, respectively. As a percentage of the legal minimum wage, variations range from 48.2 per cent

Table 4.9 Evolution of the regional expenditure on minimum income schemes

Autonomouscommunities

Actual spending (million of � )

1996 1997 1998 1999 2000

Andalusia 24.431 11.756 17.760 17.760 27.532

Aragon 2.885 2.843 2.614 2.560 2.596

Asturias 8.072 9.129 9.857 9.568 10.632

Balearics – – 1.587 0.974 1.082

Basque Country 41.320 42.437 42.197 44.817 53.238

Canaries 7.212 6.653 6.888 8.186 8.024

Cantabria 0.445 1.833 2.831 3.780 3.456

Castile-LaMancha

4.375 2.416 2.777 2.476 1.947

Castile and Leon 8.132 8.198 8.114 8.468 8.763

Catalonia 25.531 30.297 34.143 34.143 36.632

Extremadura 1.262 908 3.666 2.386 3.119

Galicia 9.604 10.578 11.696 13.150 12.699

Rioja 0.319 0.343 0.379 0.325 0.337

Madrid 20.777 22.832 24.936 24.966 24.912

Murcia 1.346 1.160 1.004 0.962 0.950

Navarre 2.458 2.747 3.005 3.486 4.075

Valencia 6.924 7.633 9.226 10.115 10.019

Total 165.092 161.762 182.678 188.123 210.012

Sources: Elaboration from data provided by the Dirección General de Acción Social, del Menor y de la Familia(Spanish Ministry of Labour and Social Affairs).

NoteBudget expenditure amount corresponds to the expenditure of the minimum income payments, except for Asturias

which also includes social wage expenditure and training.

Table 4.10 Beneficiaries of regional minimum income schemes (households)

Autonomouscommunities

Number of households

1996 1997 1998 1999 2000

Andalusia 10,603 7,907 7,284 9,914 15,962

Aragon 1,112 951 793 968 1,396

SPAIN: POVERTY AND “SAFETY NETS” 151

Autonomouscommunities

Number of households

1996 1997 1998 1999 2000

Asturias 1,511 1,761 1,067 1,022 1,512

Balearics – – 758 604 570

Basque Country 16,052 16,472 16,190 15,804 16,550

Canaries 3,096 3,153 5,690 6,689 5,358

Cantabria 490 320 1491 2,173 2,340

Castile-LaMancha

2,116 1,619 1,027 920 813

Castile and Leon 3,306 3,052 2,957 2,937 2,814

Catalonia 8,372 9,678 9,677 9,672 9,726

Extremadura 671 443 2365 946 1,351

Galicia 5,003 5,353 5,647 4,292 4,156

Madrid 7,815 7,878 8,934 8,304 7,855

Murcia 532 571 379 327 3,478

Navarre 1,503 1,646 1,770 1,264 1,820

Rioja 282 266 233 197 179

Valencia 3,713 2,644 4,404 3,470 2,565

Total 66,177 63,714 70,666 69,503 78,445

Sources: Elaboration from data provided by the Dirección General de Acción Social, del Menor y de la Familia(Spanish Ministry of Labour and Social Affairs).

NoteThe number of households in the Programme corresponds to the number of households with access to minimum

income schemes except for Asturias which also includes social employment expenditure and training andAndalusia (year 1996) which includes training.

to 64.3 per cent, and from 82.9 per cent to 110.6 per cent, as a percentage of the mean non-contributorypension of the Social Security (these are prorata gross percentages according to monthly payments).

All programmes offered lower amounts to support a life-style which could be financed were beneficiariesto be in the formal labour market earning the minimum wage, and had a rather similar level of protection asthat provided by the contributory Social Security. In general, such amounts do not reflect wide regionaldisparities, although in order to assess the real protecting intensity family and other supplements awarded tocover specific needs should also be taken into account, as is the case of the Basque Country (Sanzo 2002).

Likewise, the total number of households that received minima benefits in the year 2000 provides a goodindicator of coverage.61 This indicator shows a greater degree of disparities: while in the Basque Countrybenefits were received by 2.64 per cent of the total Basque households, percentages of between 1 per centand 1.5 per cent corresponded to Canary Islands, Cantabria, Murcia and Navarre, and the rest of theComunidades Autónomas hardly reached 1 per cent of all households.

Expenditure figures are also a good measure to assess both regional effort and generosity in financingthese programmes. Once again, the special fiscal arrangement in the Basque Country allowed thisComunidad Autónoma to afford up to 1.03 per cent (� 53.2 million) in its 2000 Budget for funding theminimum income programme. A group of regions including Andalusia, Catalonia and Madrid also devote

152 ANA ARRIBA AND LUIS MORENO

sizeable amount to these programmes. In some others, however, the budgetary effort is rather modest: 0.9per cent in the case of Asturias and no more than 0.1 per cent in regions such as Murcia or La Rioja.

Tables 4.9 and 4.10 reproduce expenditure patterns and number of beneficiaries in the second half of the1990s. If we compare figures over this period a sustained increase is noticeable. On observing annualmodifications, however, a general pattern of change for all of them cannot be inferred. Figures ofexpenditure flows and number of beneficiaries do not show a common pattern either (e.g. the BasqueCountry and Murcia). Some authors have contrasted these changes with relation to employment rates,demographic transitions and institutional variations regarding these programmes (Ayala 2001).

Variations concerning institutional and political inputs are the ones which most condition programmes’output and outcomes. Throughout the 1990s a process of adjustment and consolidation for mostprogrammes took place, which was also subject to political volatility. Programmes were implemented forreasons of legitimacy by the new mesogovernments in the general process of decentralization in Spain. Infuture they could also suffer from a certain degree of discretionary politics.

Comparisons between programmes must be carried out according to commensurable criteria. Oncontrasting the impacts made by regional minimum income programmes, those territorial imbalances anddisparities prior to their implementation must be taken into account in the first place. Any evaluativeexercise must consider the starting situation of the regions regarding their relative position concerningpoverty and wealth, citizens’ purchasing power and spending capacities. In particular, the maintenance orotherwise of traditional practices of family micro-solidarity is to be regarded as a fundamental indicator inthe fight against poverty in Mediterranean Spain.

The Comunidades Autónomas—and their mesogovernments—have benefited in terms of politicallegitimization as a consequence of the implementation of the Rentas Mínimas de Inserción. Programmes’visibility was maximized by the fact that, in the beginning of the process of implementation, the financialimplications of new minimum income schemes were not too dear for the regional treasuries. Besides, aclear message of policy innovation and political aggiornamento underlined their dynamic commitment tocarry out the weaving of safety nets of social protection in Spain. However, it remains to be seen whetherthese programmes will continue to be a priority for the regions. The Comunidades Autónomas may facea not-too-distant future situation of either requesting co-funding from the central government or containingthe coverage scope of their benefits.

All things considered, the impact of these ab novo programmes of minimum income has had a dramaticeffect in the debate about the completion of a safety net in Spain. According to estimates provided by theMinistry of Labour and Social Affairs, in 2000 there were 202,221 beneficiaries in the whole of Spain,including dependent family members (around 0.5 per cent of the Spanish population) (MTAS 2001b).

In all future scenarios, the action by the regional and local governments will be of decisive importancefor the sustainability of the system of social protection in Spain and, in particular, for the maintenance ofeffective programmes against poverty and social exclusion. Indeed, the three-layer institutional interplay isa structuring variable, which pre-determines to a great extent the diverse nature of welfare outcomes incontemporary Spain.

Conclusion

The public safety net in Spain is fragmented and composed mainly of national and regional benefits for theneedy. The contributory realm is the main source of income for poor and excluded beneficiaries, but socialassistance has gained in importance in the past decades. Despite its fragmentation, an organizing rationalecan be identified concerning the total policies tackling poverty and exclusion in Spain. Of particular interest

SPAIN: POVERTY AND “SAFETY NETS” 153

is the proactive approach taken by the Spanish regions, or Comunidades Autónomas, in the implementationof new benefits for low-income citizens.

Regions in Spain have fully exercised their constitutional right to home rule in the general framework ofEuropean subsidiarity. The role played by the Spanish mesogovernments of the Comunidades Autónomas inthe construction of public safety nets has been crucial in terms of policy innovation, and could be used as areference for developments in other Southern European countries.

Nevertheless, future scenarios for the regional programmes are uncertain. If it is true that themesogovernments have been able to integrate social services and social assistance policies into a commonregionally-based network of provision, the lack of financial resources and the ever-latent risk ofexacerbating inter-regional inequalities in welfare provision are to be underlined. Until now, the “mimesiseffect” among the Comunidades Autónomas has proved to be an effective barrier against opendiscrimination among them, and a very effective de facto equalizer of policy output.

The outcomes of regional polices in the fight against poverty and social exclusion need, in any case, to beassessed on a medium-term perspective. Longer time-series for analyses are, thus, required to validate theSpanish case of welfare decentralization. The one lesson to be drawn from the Spanish experience is theattempt to make effective both principles of Europeanization: territorial subsidiarity and democraticaccountability.62

Despite its institutional demotion (from an institutional status as independent Ministry of Social Affairs toan integrated part of a larger Ministry of Labour and Social Affairs), ministerial central officials dealingwith poverty and social exclusion matters have taken the initiative in the co-ordination of the workings forestablishing the 2001–03 Spanish NAPSI (National Action Plan for Social Inclusion). Problems with itseventual implementation mainly concerned its financing. All participants involved in the formulation of thePlan put their wholehearted support behind it. However, it was not clear whether the central governmentwould manage the contribution of “fresh” funds to accomplish the estimated increases at an annual rate of 8per cent.

Some doubts have been cast as to whether the role of the Spanish central government would be merelythat of coordinating the aggregation of funds already devoted by public and private institutions (profit andnon-profit) to promote social inclusion. Let us remember that in federalized Spain the task of bringingtogether a wide range of governmental and societal stakeholders cannot simply be achieved by displaying acentral, top-down, hierarchical harmonization of programmes nation-wide. However, the very existence ofthe NAPSI can be assessed as an important attempt at intergovernmental co-ordination as it provides thevery first synthetic overview of the fight against social exclusion in Spain.

Acknowledgements

This chapter is based on previous texts developed by both authors during the research activities carried outfor the European project, FIPOSC (Fighting Poverty and Social Exclusion in Southern Europe, ResearchDirectorate General, European Commission, HPSE-CT-2001–60020). The authors express their gratitude toTeresa Buil for her assistance in data collection and the gathering of diverse materials necessary for thediscussion in this chapter. We also thank the Spanish Secretary of State for Education and Universities(PR2002–0200) and the Region of Madrid (Dirección General de Investigación de la Consejería deEducación, Programa de Becas Posdoctorales) for financial support.

154 ANA ARRIBA AND LUIS MORENO

Notes

1 These “superwomen” have been able to reconcile demanding unpaid caring work in the household and theirincreasing professional activities in the paid labour market. Cohorts of women in the age group 40–64 are, grossomodo, representatives of this type of “superwoman”. For an analysis of the case of Spain, see Moreno (2004).

2 In 1995, 62 per cent of Spaniards were of the opinion that “The State is responsible for the welfare of each andevery one of the Spanish citizens and has the duty to help them to solve their problems”, as compared to 15 percent who considered that “The State is only responsible for the welfare of the least-favoured citizens and has theduty to help them to solve their problems”, and 16 per cent that “The citizens are responsible themselves for theirown well-being and have the duty to sort out their own problems”. Such percentages have been similarthroughout the 1990s (Del Pino 2001).

3 Note that social services were already a domain for private intervention. The new scenario is precisely thebuilding of public networks within which the third sector plays a subsidiary—in many cases publicly subsidized—role.

4 Note that regional expenditure increased its share in total Spanish public spending from 3 per cent in 1981 to asmuch as 33 per cent in 2000 (MAP 1997).

5 However, municipalities were the single institutions which received most requests for help by poor families in allfour types of poverty, as defined according to income percentages regarding the average national disposableincome (extreme, lower than 15 per cent; grave, between 16 per cent and 25 per cent; moderate, from 26 per centto 35 per cent; precarious, between 36 and 50 per cent). The corresponding figures in 1994–96 were,respectively, 33 per cent, 31.3 per cent, 35.5 per cent, and 39.5 per cent of the total (Rodríguez-Cabrero 1998).

6 Or NAPincl, as expressed in documents of the European Commission.7 Health services were delivered by both public and private institutions in agreement with the social security

system. This latter arrangement secured the incomes of medical doctors, who in most cases also worked in theprivate sector. The pharmaceutical industry profited abundantly as well. In turn, the quality of the public healthservice was very poor (Moreno and Sarasa 1993).

8 With the implementation of the 1959 Stabilization Plan public expenditure was reduced, the peseta currency wasdevaluated, and investment controls were relaxed. Foreign holdings of up to 50 per cent in Spanish companieswere permitted. The results of the desarrollismo were impressive, with GDP growing by an annual 7 per centbetween 1960 and 1974.

9 Responsibilities in the provision of social services for municipalities of more than 20,000 inhabitants had alreadybeen established by the Basic Law for Local Government passed by the Spanish Parliament in 1982.

10 The Ministry was formed by the General Directorate of Social Action, the INSERSO, and the Institutes for theWomen and the Youth, and the Board for the Education and Care of the Disabled. It took also the responsibilityof supervising the activities of NGOs, such as the Red Cross, the Organization of the Blind, and other non-profit-making private charities.

11 The Basque government did not join this general agreement because it did not accept categorical financing.12 However, the legal minimum wage decreased from 77.5 per cent of the per capita GDP, in 1980, to 52.4 per cent,

in 1992. Note that ten years later, the minimum wage (Salario Mínimo Interprofesional) in Spain was fixed at� 516 per month, the third lowest in Europe. However, as compared with other EU countries, the number ofworkers with a minimum wage was very small, around 2 per cent or 165,000 employees.

13 Following its parliamentary approval, on 9 October 1996 the main trade unions (CCOO and UGT) and thegovernment signed the “Agreement for the Consolidation and Rationalization of the Social Security System”(Acuerdo Consolidación y Racionalización del Sistema de Seguridad Social). The Law 24/1997 on“Consolidation and Rationalization of the Social Security” (Ley de Consolidación y Racionalización de laSeguridad Social) was based on the previous “Proposal” and “Agreement”. The trade unions had given theirconsent until 2000. In April 2001 a new “Agreement for the Improvement and Development of the System ofSocial Protection” was signed by government, the employers’ associations (CEOE and CEPYME), and theCCOO. The other main trade union, UGT, disagreed with some of the contents of the text and refused to sign.

SPAIN: POVERTY AND “SAFETY NETS” 155

14 The process of separating the sources of financing was not fully achieved during the period 1997–2002, althoughthe patterns for both contributory and non-contributory realms of social spending were fixed. As a result of thegradual separation of the sources, budget surpluses in the contributory social security system (social insurance)have been achieved since 1999. These surpluses have been put into a Reserve Fund and are earmarked to coverany possible financing shortages in the future, as well as to strengthen the long-term viability of the “pay-as-you-earn” public pension system. Fund money amounted to � 8 billion on 4 July 2003 (Rodríguez-Cabrero et al.2003).

15 In April 2002, the government made public its proposals to reform unemployment provisions, which have beenhighly contested by the trade unions and the opposition parties. Among other conditions, the proposals were setto require beneficiaries to accept jobs distant up to 50 kilometres from his/her place of residence. Likewise,beneficiaries would lose their right to the benefit after refusing for the third time to take on a job offered to him/her considered as suitable by the state employment or manpower office (INEM).

16 Law 12/2001 makes provisions for tax breaks in social contributions to newly hired employees in needysituations (e.g. beneficiaries of minimum income schemes). It also awards the status of insertion companies,alongside some fiscal privileges, to those hiring low-income workers. Note that despite the support provided bysocial agents to the establishment of these “productive” businesses, no legal provisions have so far beenimplemented to regulate their activities.

17 As pointed out by Ruíz-Huerta et al. (2000), the Mínimo was “original” in part as there were previous fiscalschemes which had also taken into consideration non-taxable income minima. However, it was innovative as amechanism for the evaluation of the resources that the various types of households required to cover their basicneeds. Analogies between the minimo and the poverty threshold are evident.

18 These policies have a markedly pro-natalist orientation and are a reaction to the very low birth rate in Spain andthe increasing participation of women in the labour market.

19 At present, and further to the deductions related to the Mínimo Personal y Familiar, other tax breaks are availablefor working mothers with children under 3 years of age. However, discounts for school material for childrenbetween 3 and 15 years have been phased out.

20 Articulated not only by nationalist and regionalist parties, but also by the increasingly important regional andfederated branches of the main Spanish political formations (PP, PSOE, and IU) (Arriba 1999). Internalprocesses of power accommodation within the Spanish parties have also greatly contributed to the internalizationby Spaniards of the federalization of politics and policymaking (Moreno 2001b).

21 Three main “basic surveys” were carried out in 1973/74, 1980/81 and 1990/91 with samples of more than 20,000households. Since 1985 the Continuous Survey on Family Accounts (Encuesta Continua de PresupuestosFamiliares, ECPF), has been carried out as a three-month survey with a smaller sample: out of the number of 3,200 households, one-eighth of them are renewed every term. In 1997, the ECPF sample increased up to 8,000households and the main “basic surveys” were discarded. Note that the assessment of the data produced by the“basic surveys” was conditioned by the lack of periodicity: results were representative of the period when theyhad been produced but they could not provide information on time changes and the dynamics of inequalities.With the implementation of the ECPF surveys this problem of periodicity was partially solved. But the ECPFsurveys have provided less information and, thus, reliability. This is due to their smaller samples and to theimpossibility of disaggregating information territorially. This latter problem was somewhat mended with anincrease in the size of the sample in 1997. Another problem faced by the EPF, as in other surveys of this kind, isthe “underestimation” of monetary data, in particular those referring to income. To all these problems one furthershortcoming should be added: the delay in the publication of the ECPF micro data, which has so far renderedvery difficult the elaboration of assessment studies and further research on poverty based on these data sources(RuízHuerta and Martínez 2000).

22 This is a fixed panel of 8,000 Spanish households (70,000 EU-wide), which provides longitudinal informationfrom 1994 to 2000. The research design reduces considerably the “underestimation” problems concerning incomedata. It also collects information relevant for studies on poverty of a multi-dimensional nature, as well as data oneducation and employment which may feasibly improve the consideration of explanatory and interpretative

156 ANA ARRIBA AND LUIS MORENO

factors. However, among its shortcomings two can be pointed out: (i) the lack of information on non-monetaryincome and on consumption patterns; and (ii) the sample size makes it impossible to work out reliableestimations at the regional level (as happens with ECPF). The ECHP started in 1994 but the data produced werepublished after a substantial delay.

23 The first of these was carried out at the beginning of the 1980s. Between 1994 and 1997, provincial surveys werealso done and their results were analysed at both provincial and national levels. One of the main shortcomings ofthese surveys concerns the selection of the sample (which is made by means of a “filter” question on familyincome carried out in a personal interview in the respondent household).

24 Comparisons between the results provided by the various sources have to be carefully established due tocommensurability problems. Caution is also advisable in the analyses of the differences produced within eachsource because of the error margins of these surveys.

25 As the FOESSA Reports, which were carried out under the auspices of Caritas (1967, 1970, 1975, 1983).26 Considering the threshold of 50 per cent of mean equivalized income, the poverty rate for 1973–74 according to

EPF was around 20 per cent of the total population. For the threshold of 25 per cent, the poverty rate was slightlyhigher than 4 per cent (Ayala and Renes 1998).

27 Accordingly, ECPF data in 1998 established the poverty rate in Spain (population under the 50 per cent of themean equivalized income) at around 12 per cent (Ayala et a/. 2002).

28 Most of the members of this group are people over 65 years of age.29 Let us remember that the use of OECD equivalence scales tends to “inflate” poverty rates as the size of

households also increases.30 For an analysis of the effects on poverty and inequality produced by an accumulation of the different sources of

income (employment, capital, contributory benefits, assistance benefits and private assistance), see Carabaña andSalido (2001).

31 The poverty gap index (FGTl) is the income shortage of the poor with reference to the poverty line. It can beinterpreted as the amount needed to be transferred to the poor so that their resulting income can reach the povertyline.

32 Note that access to these benefits is only possible if there have been a previous contributory trajectory of thebeneficiary, despite the possibility that this might have been insufficient to sustain the entitlement.

33 Unemployment reform proposals put forward by the Spanish Government in the year 2002 also included thegradual disappearance of the agrarian subsidies (PER-REASS), and their substitution by a contributory regimefor all eventual agrarian labourers similar to the general provisions for the unemployed.

34 Note that minimum wage amounts (salario mínimo interprofesional) are referred to full-time jobs. In the case ofsome regional minimum income programmes, and due to added amounts as family supplements, the total benefitamount can be higher than that of the minimum wage.

35 This is to be understood as the income of the household, or common unit of residence.36 Data in this section have been collected from the Spanish NAPSIs, Planes Nacionales para la Inclusion del

Reino de España 2001–2003 y 2003–2005 (MTAS 2001a; 2003), as well as information provided viva voce ininterviews with two members of the Plan Secretariat in February 2002. We are grateful for their collaboration.

37 Five officials of the General Sub-Directorate of Programmes of Social Services carried out activities of the PlanSecretariat. No extra budgetary provision was made available to the Secretariat, nor were officials from othergovernmental bodies involved in the workings. Nothing can be compared to the resources devoted in other similarprocesses, such as those related to Employment Plans. Let us say that some external support was hired for theprocessing of data and information.

38 See Appendix III, MTAS (2001a, 2003).39 The Spanish Economic and Social Council has the status of a government advisory body, which allows the

economic and social partners to participate in economic and social policy decision-making. The Council mustissue a mandatory opinion on preliminary drafts of State Acts and draft Royal Statutory Instruments regulatingsocioeconomic and labour issues, as well as draft Royal Decrees considered to be of special importance by the

SPAIN: POVERTY AND “SAFETY NETS” 157

government. Additionally, it performs an important job in promoting dialogue, which leads organizations thatlegitimately defend clearly distinct interests to find common points and share growing spheres of opinion.

40 The support from the NGOs was channelled through the Spanish Council of Institutions for Social Action (ConsejoEstatal de Entidades de Acción Social, which was established on 19 February 2001). Let us remember that thoseorganizations usually participate in programmes financed by general taxation regarding projects of social action.They are funded by 0.5 per cent of those income tax payments so explicitly targeted by taxpayers in their annualincome tax statements (the other option is to transfer such money to the Catholic Church). In 1998, � 89.9 millionwere collected under this 0.5 per cent entry, an amount which is an important source of funding for many of thesenon-profit-making organizations.

41 Including the Spanish Ombudsman Office (Defensor del Pueblo).42 A workfare priority already included in the Spanish National Action Plan for Employment was labour activation

for socially excluded groups. The Employment Plan regarded the implementation of specific policies to facilitateinsertion in the labour market.

43 Many actions and measures are reproduced under different sections. For instance, mention of ApprenticeshipProgrammes (Escuelas Taller y Casas de Oficios) are found in many epigraphs within the Annex of theevaluation document. Among the innovations, the 2003–05 Plan incorporates as a strategic aim the achievement of a quantitativedecrease in the poverty rate. New measures, such as those related to justice matters and the involvement of thetargeted groups are also taken into account, together with trend analyses and a selection of benchmarking cases.Yet again, the problems affecting data sources on poverty and inequality in Spain can be emphasized. Consistentsources and databases would be useful to design indicators for consistent evaluation. Until 1999, data providedby the European Community Household Panel somewhat covered the insufficiencies already mentioned in theprevious section of this chapter.These Plans incorporate as core policy instruments the regional minimum income schemes for insertion (RentasMínimas de Inserción).Once again, the importance of the decentralized structure of the Spanish “State of Autonomies” (Estado de lasAutonomías) can be emphasized. As a consequence of the regional responsibilities on social assistance and socialservices matters, the Spanish NAPSI locates the main responsibility for the fight against exclusion at the regionallevel and, in so doing, action is somewhat dependent on the action developed by the TAPSIs.Caritas produced a comprehensive Plan proposal (Caritas 2001). This document was highly critical towards theexisting policies of social inclusion in Spain, but many of its proposals were incorporated into the final version ofthe Spanish NAPSI. This lack of comprehensive proposals by the Spanish government greatly explains whyONGs such as Caritas took a leading role in the elaboration of the Plan (Razón y Fé 2001).In some cases it can be deduced from the measures put forward by the latest NAPSI: the achievement of anintergovernmental agreement for a common scheme of minimum income guaranteed indicates still appears as animportant “pending matter”.The identification of these citizens is not an easy task concerning the ambivalence in the concept of socialexclusion.Although not all the contributions of these associations were incorporated into the document of evaluation of theNAPSI.For instance, the counting of households that can have access to the RMIs programmes and the programmeactivities taken into account (e.g. insertion activities, benefits for specific needs or workfare programmes) mayshow dissimilarities.Results differ from the data reproduced in Table 4.6, which was elaborated from detailed analyses not only of thelegislative and budgetary provisions of each programme, but also of implementation and functioningarrangements. Differences between these two kinds of data mainly relate to: (i) the relative “weight” ofcomplementary benefits according to family size in the protection intensity; (ii) differences in the regionalpoverty rates in order to asses programme coverage and “generosity”; and (iii) the legal restrictions and thediscretionary levels and budgetary limits.

158 ANA ARRIBA AND LUIS MORENO

According to 1995 data, the mean non-financial per capita autonomous spending carried out by Catalonia andGalicia was � 1,373, which compared to � 2,509 in the Basque Country.

55 In practice, this procedure is rather problematic. On the one hand, the legal beneficiary can make an abusive andpersonal use of the benefit. On the other, the beneficiary could also be overburdened with insertion obligationssigned on accepting the subsidy. (In many cases he/she is the only one in contact with the responsible socialworker.) In some qualitative studies it has been noted that contacts with social workers is mainly carried out by awoman of the family, despite the fact that the entitlement is generally awarded to “the man of the family”. In theparticular case of Aragon, the entitlement can be shared and the cash benefits can accordingly be divided betweenadult members of the family. Problems of a practical nature and “perverse” effects in the implementation of theseprogrammes have been sorted out in many cases in a contingent manner (Serrano and Arriba 1998).

56 This is the case, for instance, of some extended families with elderly members entitled to old-age pensions. Adhoc adjustments in these cases have been made together with a certain relaxation of legal requirements.

57 Note that, despite their non-contributory nature, the Social Security is responsible for the payment of thesepensions. Main problems associated with such age limits are: (i) amounts provided by non-contributory pensionscan be lower than that of rentas mínimas; and (ii) non-contributory benefits of citizens over 65 years of age withdependent family members may be clearly insufficient.

58 Exception made for those cases of beneficiaries aged under 25 with dependent family members who areincluded.

59 Some later reforms have aimed at achieving a minimum income scheme in sensu strictu, with no direct linkage toinsertion, very much along the lines of late developments of the French RMI.

60 Data reproduced was directly facilitated to the General Directorate by the Comunidades Autónomas. We shouldpoint out that interpretations on these data must be carried out cautiously because if in some cases they refergenuinely to minimum income, in some others (e.g. Andalusia and Asturias) they include disaggregate figurescorresponding to benefits for socially protected employment or to finance insertion projects. In some other cases,figures add up the aggregate number of all beneficiaries who have joined the programme in a year (e.g. BasqueCountry), but other programmes take into account the number of beneficiaries in a given month (e.g. Madrid).

61 Current data on their incidence on regional poverty rates are not available, something which has prevented usfrom calculating coverage rates of the targeted population. Further to this, the data of the General Directorate forSocial Action on minimum income beneficiaries are based on estimations. Consequently, the option taken hasbeen to calculate coverage rates of households in the Comunidades Autónomas.

62 Let us remember that according to the views of the European Commission, the new European governance is notconsidered the “exclusive” responsibility of EU institutions. Neither national governments nor nationalparliaments are regarded as being the sole actors of European governance. Local authorities and the regions areregarded as the decisive emerging actors (Commission 2000).

References

Agranoff, R. (1993) “Las relaciones intergubernamentales y el Estado de las Autonomías”, Política y Sociedad, 13:87–105.

Aguilar, M., Gaviria, M. and Laparra, M. (1995) La caña y el pez. El salario social en las Comunidades Autónomas1989–1994, Madrid: Fundación Foessa.

Aliena, R. (1991) “RMI, le Gouvernement espagnol a contre-courant”, Revue Française des Affaires Sociales, 45:81–97.

Arriba, A. (1999) “Rentas mínimas de inserción en España: Procesos de implantación y dinámicas sociales”, PhD thesis,Universidad Autónoma de Madrid.

Arriba, A. (2002) “Procesos de implantación de políticas de rentas mínimas de inserción en España”, in L.Moreno (ed.),Pobreza y exclusión: la “malla de seguridad” en España, Madrid: CSIC.

Ayala, L. (2000) Las rentas mínimas en la reestructuración de los estados de bienestar, Madrid: CES.

SPAIN: POVERTY AND “SAFETY NETS” 159

Ayala, L. (2001) “El ingreso madrileño de integración: una valoración desde el conjunto de las ComunidadesAutónomas”, in M.Aguilar, M.Laparra, and B. Pérez (eds), La exclusion multidimensional en el espacio urbano:Investigaciones de base para la elaboración del Plan de Lucha contra la Exclusión Social en la Comunidad deMadrid, Madrid: Comunidad de Madrid, pp. 55–80.

Ayala, L., Martínez, R. and Ruíz-Huerta, J. (2001) “La descentralización territorial de las prestaciones asistenciales:Efectos sobre la igualdad”, Madrid: Instituto de Estudios Fiscales, P.T. 16/01. Available online: http://www.ief.es(accessed June 2003).

Ayala, L. and Renes, V. (1998) “El estudio de la pobreza en España”, in EDIS et al., Las condiciones de vida de lapoblación pobre en España, Madrid: Fundación FOESSA.

Ayala, L., Ruíz-Huerta, J., Martínez, R. and Sastre, M. (2002) Perfil de la población española en pobreza y riesgosocial: situación y tendencias a partir del Panel de Hogares de la Union Europea (1993–1997) y la EncuestaContinua de Presupuestos Familiares, Madrid: MTAS, unpublished report.

Beltrán, M. (1992) El régimen jurídico de la acción social pública, Oñati: IVAP.Cantó, O., del Rio, C. and Grandín, C. (2002) “La evolución de la pobreza estática y dinámica en España en el periodo

1985–1995”, Madrid: Instituto de Estudios Fiscales, P.T. 24/02. Available online at http://www.ief.es/Publicaciones/ papeles/02/pt2402.pdf (accessed June 2003).

Carabaña, J. and Salido, O. (2001) “Fuentes de renta, desigualdad y pobreza de individuos y hogares (España, 1993)”,in L.Moreno (ed.), Pobreza y exclusion: La “malla de seguridad” en España, Madrid: CSIC, pp. 107–152.

Cáritas (2001) Plan Nacional para la inclusion social Propuesta de Cáritas, Documento de Trabajo, June, Madrid:Cáritas Española.

Casado, D. (1990) Sobre la pobreza en España 1965–1990,Barcelona: Hacer.Casado, D. et al. (1994) “Acción social y servicios sociales”, FOESSA, V Informe Sociolágico sobre la situación en

España: Sociedad para todos en el año 2000. Madrid: Fundación FOESSA.Castles, F. (2001) ‘The future of the welfare state: crisis myths and crisis realities”, paper presented at the ISA RC19

Conference, “Old and New Social Inequalities: What Challenges for Welfare States?”, 6–9 September, Universityof Oviedo.

CES (Consejo Económico y Social) (2001) La pobreza y la exclusion social en España: Propuestas de actuación en elmarco del plan nacional para la inclusion social, Madrid: CES.

Commission of the European Communities (2000) “Shaping the New Europe”, Strategic Objectives 2000–2005,Brussels: COM (2000) 154 final.

Cruz Roche, I. (1994) “La dinámica y estructura de la universalización de las pensiones”, V Informe Sociológico sobrela Situación Social en España, Madrid: Fundación FOESSA.

Cruz Roche, I., Desdentado, A. and Rodríguez-Cabrero, G. (1985) Política Social y Crisis Económica, Madrid: SigloXXI.

Del Pino, E. (2001) “Las expectativas ciudadanas sobre la acción pública: el proceso y los resultados de las políticas debienestar en España”, paper presented at V Spanish Congress of Political Science, 26–27 September.

Eardley, T., Bradshaw, J., Ditch, J., Gough, I. and Whiteford, P. (eds) (1996) Social Assistance in OECD Countries:Synthesis Report, London: HMSO.

EDIS et al. (1998) Las condiciones de vida de la población pobre en España, Madrid: Fundación FOESSA.EU (2002) Social Protection in Europe, 2001, Brussels: European Commission.Ferrera, M. (1996) “The ‘Southern Model’ of welfare in social Europe”, Journal of European Social Policy, 6:17–37.Ferrera, M., Matsaganis, M. and Sacchi, S. (2002) “Open coordination against poverty: the new EU ‘Social inclusion

process’”, Journal of European Social Policy, 12:227–239.García Serrano, C., Malo, M.A. and Toharia, L. (2001) La pobreza en España: Un an álisis crítico basado en el Panel

de Hogares de la Union Europea (PHOGUE), Madrid: MTAS.Guillén, A.M. (1992) “Social policy in Spain: from dictatorship to democracy (1939–1982)”, in Z.Ferge and

J.E.Kolberg (eds), Social Policy in a Changing Europe, Boulder, CO: Westview Press.Guillén, A.M. and Matsaganis, M. (2000) “Testing the ‘social dumping’ hypothesis in Southern Europe: welfare

policies in Greece and Spain during the last 20 years”, Journal of European Social Policy, 10:120–145.

160 ANA ARRIBA AND LUIS MORENO

Laparra, M. and Aguilar, M. (1997) “Social exclusion and minimum income programs in Spain”, in MIRE, ComparingSocial Welfare Systems in Southern Europe, vol. 3, Florence Conference, Paris: Mission Recherche etExperimentation (MIRE), pp. 515–535.

MAP (1997) Estudio sobre reparto del gasto público en 1997 entre los distintos niveles de administración, Madrid:Ministerio de Administraciones Públicas.

Ministerio de Asuntos Sociales (1989) “Informe sobre diferentes prestaciones sociales en la CEE y España”,unpublished report.

Moreno, L. (2000a) “The Spanish development of Southern European welfare”, in S.Kuhnle (ed.), Survival of theEuropean Welfare State, London: Routledge, pp. 146–165.

Moreno, L. (2000b) Ciudadanos precarios. La “ultima red” de protección social, Barcelona: Ariel.Moreno, L. (2001a) “Spain, a Via Media of welfare development”, in P.TaylorGooby (ed.), Welfare States Under

Pressure, London: Sage.Moreno, L. (2001b) The Federalization of Spain, London: Frank Cass.Moreno, L. (2002a) Pobreza y exclusion: la “malla de seguridad” en España, Madrid: CSIC.Moreno, L. (2002b) “Decentralization in Spain”, Regional Studies, 36:399–408.Moreno, L. (2003) “Europeanisation, mesogovernments and safety nets”, European Journal of Political Research, 42:

185–199.Moreno, L. (2004) “Spain’s transitions to New Risks: a farewell to ‘superwomen’”, in P.Taylor-Gooby (ed.), New Risks

and New Welfare in Europe: The Transformation of the European Welfare State, Oxford: Oxford University Press,pp. 137–160.

Moreno, L. and Arriba, A. (1999) Welfare and Decentralisation, Working Paper EUF 99/8, Florence: EuropeanUniversity Institute.

Moreno, L. and Sarasa, S. (1993) “Genesis y desarrollo del Estado del Bienestar en España”, Revista Internacional deSociolog�a, 6:27–69.

MTAS (2001a) Plan Nacional de Acción para la Inclusion Social del Reino de España, Junio 2001–Junio 2003,Madrid: Ministerio de Trabajo y Asuntos Sociales.

MTAS (2001b) Anuario de Estadísticas Laborales y de Asuntos Sociales, 2000, Madrid: Subdirección General deEstadísticas Sociales y Laborales, MTAS. Available online at http://www.mtas.es/Estadisticas/anuario00 (accessedMarch–June 2002).

MTAS (2003) Plan Nacional de Acción para la Inclusion Social del Reino de España 2003–2005, Available online http:(accessed August 2003).

Razón y Fé (2001) “Plan Nacional de Acción para la Inclusion Social”, Razón y Fé, 1233:27–32.Rhodes, M. (ed.) (1996) “Southern European welfare states”, South European Society and Politics 1 (3), Special Issue:

1–300.Rodríguez-Cabrero, G. (1989) “Orígenes y evolución del Estado del Bienestar español en su perspectiva histórica: Una

vision general”, Política y Sociedad, 2:79–87.Rodríguez-Cabrero, G. (1994) “La política social en España: 1980–92”, V Informe Sociológico sobre la Situación

Social en España, Madrid: Fundación FOESSA.Rodríguez-Cabrero, G. (1998) “Servicios sociales y pobreza” en EDIS et al., Las condiciones de vida de la población

pobre en España, Madrid: Fundación FOESSA.Rodríguez-Cabrero, G., Arriba, A. and Marbán, V. (2003) Spain’s National Report on Policy Maps of Welfare Reform,

Madrid: Working Paper UPC 03–11. Available online at http://www.iesam.csic.es/doctrab.htm (accessedSeptember 2003).

Ruíz-Huerta, J., Martínez, R. and Ayala, L. (2000) “El mínimo personal y familiar en el IRPF: Una valoración de sucuantía”, Hacienda Pública Española, 152:151–170.

Ruíz-Huerta, J. and Martínez, R. (2000) “El estado de las fuentes para el estudio de la desigualdad y la pobreza”, inI.Bazaga, J.A.Ramos and M.Tamayo (eds), Pobreza y desigualdad en España: Enfoques, fuentes y acción pública,Cuadernos de Gobierno y Administración 2 (special issue), Madrid: Universidad Rey Juan Carlos.

SPAIN: POVERTY AND “SAFETY NETS” 161

Sanzo, L. (2002) “La lucha contra la pobreza en Euskadi”, in L.Moreno (ed.), Pobreza y exclusion: la “malla deseguridad” en España, Madrid: CSIC.

Sarasa, S. (1993) El servicio de lo social, Madrid: Ministerio de Asuntos Sociales.Sarasa, S. and Moreno, L. (eds) (1995) El estado del bienestar en la Europa del sur, Madrid: CSIC.Serrano, A. and Arriba, A. (1998) Pobres o excluidos? El Programa del Ingreso Madrileño de Integración en

perspectiva comparada, Madrid: Fundación ArgentariaMinisterio de Trabajo y Asuntos Sociales-Editorial Visor.Serrano, A. and Arriba, A. (2002) “El Ingreso Madrileño de Integración: Revisitando sus características y algunos

tópicos”, in L.Moreno (ed.), Pobreza y exclusion: la “malla de seguridad” en España, Madrid: CSIC.Susín, R. (2000) La regulación de la pobreza: El tratamiento jurídico-político de la pobreza: los ingresos mínimos de

inserción, Logroño: Universidad de La Rioja.

162 ANA ARRIBA AND LUIS MORENO

5Portugal—a virtuous path towards minimum income?

Luís Capucha, Teresa Bomba, Rita Fernandes and Gisela Matos

Introduction1

Portugal joined the European Economic Community in 1986. In doing so, Portuguese society achieved acommon goal: to put an end to a long cycle of historical under-development that had begun 58 years beforewith the conservative and corporatist dictatorship of the “Estado Novo”,2 which only came to an end in1974 with the democratic revolution of 25 April.

While all across Europe modern welfare states were being developed, Portugal remained tied to abackward, traditionalist and under-developed economic and social organization. At the time of transition todemocracy, poverty affected more than four people in ten (M.Silva 1984) ,3 but this was no reason forpolitical concern by the authorities.

During the dictatorship, despite ambitious legislation, the social protection system did not cover themajority of the population and a national health service did not exist. Investment in education was paltry.Industrial relations were replaced by the corporative organization of the state and social movements wererepressed (Schmitter 1975; Magone 1997).

The Constitution of 1976, introduced in the wake of the 1974 revolution, established a set of social rights,which was to constitute the formal framework of the Portuguese welfare state, in many aspects now similarto the models in other European countries. Until 1986 Portugal suffered periods of great instability andsevere constraints, characterized by a sharp decrease in GDP, a chaotic productive system, a lack of theminimum financial resources to ensure the quality of the policies created, and a scarcity of institutionalresources to put them into practice. Under those conditions, social regulation was strongly supported by family-provided private welfare, and a “parallel state” emerged, strongly linked to informal economy,4 which inturn substantially limited the scope and efficacy of “official” social policies. Such facts did not prevent agreat change occurring in the country. The number of beneficiaries, the risks covered and the scope ofbenefits all increased during this period. There were also changes affecting trade unionism and the freedomof association. However, measures to fight poverty remained minimal and social action and assistance werefragmented. Accordingly, the level of poverty remained extremely high.

Following Portugal’s entry into the European Community (EC) in 1986, a rapid modernization process, witha particular impact on infrastructure and technology, was set in motion, benefiting incidentally from thefavourable international macroeconomic environment. Participation in the II European Anti-PovertyProgramme, from 1987 to 1989, was the first official step towards anti-poverty social policies. The firsttraining policies were then initiated, along with various experiments in the rehabilitation of the disabled,vocational training, and employment for groups excluded from the labour market. Moreover, the current

systems of social security and assistance (both contributory and non-contributory), health, education,training, and employment followed a process of reorganization and development.

They still face, however, some of the major difficulties involved in modernization processes. Family careservices are scarce in a society with high female employment, which means that women bear the doubleburden of maintaining a household and carrying out their professional activity; employment levels are highbut job quality, wage levels, and productivity are relatively low; in many aspects the organization of labouris anti-quated; the relative weighting of the informal economy remains high; average qualifications amongthe working population are low, as is participation in life-long learning; and the combined effects of taxevasion and associated practices, the low wages bills on which the social security contributions are leviedand the high expenditure on non-contributory benefits severely limit the financial resources available forsocial policies and mean that the amounts of social security benefits remain quite low.

In 1995 a new generation of active social policies was launched in Portugal. These policies are active in adouble sense. As in the rest of Europe, they aim to activate individuals. However, in Portugal they aim toactivate institutions as well. The significance of this is that the need to develop was felt not only in order toface the challenges of new social problems, by means of an individualized approach to citizens and byreforming policy systems, but also for the country to converge with the “acquis communautaire” in the fieldof social solidarity, and simultaneously increase economic growth to which these social policies weresupposed to contribute.

Within this framework, the Guaranteed Minimum Income (GMI) is not the only active policy measureintroduced, but is indeed a model one. This chapter positions the GMI in the process of social policydevelopment in Portugal. It begins by presenting the historical context of Portuguese social policy and therespective institutions. The evolution of poverty and inequality, labour market main features, changes infamily structures and ageing are assessed, as well as the context of recent trends in social protection andsocial inclusion strategy in Portugal. Then the GMI scheme is described, covering its design, itsimplementation, the current debate, and the main challenges to be faced. The chapter concludes with adiscussion on the future of the new generation of social policies in the framework of Portugueseparticipation in the open method of co-ordination on social inclusion.

Social protection and anti-poverty policies: historical background

Poverty was not addressed as a problem in Portugal until the mid-1980s and therefore it was never one ofthe main issues on the political agenda. Attention to the problem was mainly provoked by scientificresearch. A precursor study published in 1985 concluded that 35 per cent of family households lived inpoverty in 1980 (Costa et al. 1985).5 A study published later compared Portugal with other Europeancountries: 20.4 per cent of Portuguese households had expenditure per adult equivalent below 40 per cent ofthe national average, while, for the same year, this figure was 11.9 per cent in Spain and 11.6 per cent inGreece (Eurostat 1990).

When Portugal joined the EC in 1986, the problem gained visibility, and in June 1988 Parliamentformally discussed the issue for the first time. This was due to an initiative by the Communist Party. Thepractical consequences, besides the symbolism of being the first official debate on poverty, were nil.Debates, studies, news reports and other references have been increasing ever since, allowing a subject thatwas totally absent from policies to begin to appear and play an important role in policy-making in Portugal.

164 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Pre-25April 1974—a brief description

Whatever their particular definition, welfare states found their roots in Europe in the last decades of thenineteenth century, though they have since developed in a small number of countries in other continents aswell. Portugal took part in that process. The first legal attempt to create a comprehensive social protectionsystem occurred during the First Republic6 (see Table 5.1).

In 1919 the government passed a number of laws creating, on paper, a compulsory social insurancesystem that was fairly advanced for its time. The insurance was to be financed by contributions fromemployers and insured workers. The whole of the working population, regardless of sex or occupation, wasto be covered, on the one condition that their wages were below a certain ceiling (except for labour injuries,in which case the limit did not apply). Compulsory social insurance covered the risks of illness, accidents atwork, old age, disability and the support of dependants, in the case of death. In addition, social labourchanges were set up to employ the jobless by providing a stimulus for public works. Included in thislegislation was the creation of the Instituto de Seguros Sociais Obrigatórios,7 which would be responsiblefor regulating the system. The measures were launched with no assessment of their economic viability andno participation by employers and employees. Accordingly, the whole initiative was a complete failure asthe laws were never put into practice (Maia 1985).

Until the 1930s, there was only a rudimentary and symbolic set of welfare policies in action, some ofthem created exclusively for a small number of industrial workers and public servants. Mutualityassociations created in the nineteenth century ran some of the schemes. These associations have decreasedin number and in the breadth of guarantees offered, given the hostility they faced from the authoritarianregime. The majority of the population relied on church organizations, private charities and paternalisticfamily welfare for the provision of care.

The first legislative step of the Estado Novo to create the basis of its social policy occurred in 1933 whenthe government took the initiative of placing the existing schemes under its own control. The 1933Constitution attributed “to the state the responsibility of promoting and developing solidarity, welfare, co-operation and institutions providing mutual support”. It was subject to the basic principles of the nationallabour regulations published that same year.8

Law 1884 of 16 March 1935, following the Constitution and the Estatuto do Trabalho Nacional9 of 1933,created four categories of social protection institutions. The first category was made up of the institutions ofcorporative bodies (Caixas Sindicais de Previdência, Caixas de Previdência das Casas do Povo, Casas dosPescadores). They were compulsorily financed by contributions from employers and employees andincluded wage earners in trade and industry, farm workers, and registered seamen. The Caixas Sindicais dePrevidência (based on compulsory social insurance) covered the risks of sickness (income benefit in case oftemporary incapacity to work) and health care provision,10 invalidity, old age and involuntaryunemployment of industrial and trade employees. Later on, family allowances were also included. Foragricultural workers the compulsory scheme only covered medical care, sick pay and death allowance,provided by the Caixas de Previdência das Casas do Povo. Seamen and fishermen were only covered formedical care by the Casas dos Pescadores. The Casas do Povo and Casas dos Pescadores were both fundedby the contributions of members and occasional donations. Apart from the compulsory schemes, they couldcreate mutuality associations for the provision of small retirement pensions.

The second category of institutions consisted of the retirement and welfare funds for workers in a relationof subordination in certain companies and sectors of activity. These schemes corresponded to thecorporative welfare funds in those companies and sectors alone. They were compulsorily financed bycontributions from employers and employees and covered workers in a relationship of subordination inindustry and in

LUÍS CAPUCHA ET AL. 165

Table 5.1 Social protection in Portugal—historical overview (before 25 April 1974)

Period First Republic (1910–26) Fascist dictatorship (Estado Novo 1926–74)

First period (1933–62) Second period (1962–73)

Legal framework Laws 5636; 5637; 5638;5639 and 5640; all of 10 May1919

Law 1884 of 16 March 1935(following the Constitutionand the National LabourStatute of 1933)

Law 2115 of 1962

Organization Creation of the Institute forCompulsory Social Insurance

Creation of four categories ofsocial protection institutions,the most important being thatof the Corporative Bodies’Welfare Institutions

The government is given theresponsibility of co-ordinating social policyThe Social Council ofMinisters is createdThe four categories ofinstitution are retainedThe Federation of Welfareand FamilyAllowance Funds and theNational PensionFund are set upOrganization of the systemon a district basiscommences

Financing Contributions fromemployers and insuredworkers

Contributions fromemployers and insuredworkers

Contributions fromemployers and insuredworkers

Method Capitalized funds Capitalized funds Combined system (fundedand pay-as-you-go)

Main risks covered Illness; accidents at work;old age; disability;surviving dependants’benefitSocial labour exchangesfor the unemployed,stimulated by public works

Sickness, disability, oldage, involuntaryunemployment (later,family allowances),medical care and death—incorporative institutions

Sporadic improvements tothe earlier system: in thespecial family allowancescheme, the special welfarescheme, survivingdependants’ pensions,occupational diseasepensions, sickness benefit,death benefit and assistanceto sick or handicappedchildren allowance

Categories of peoplecovered

The active population, ofboth sexes, on thecondition that wages werebelow a certain ceiling(except for accidents atwork)

Wage-earners in trade andindustry; farm workers andregistered seamen;employees in a relationshipof subordination in industryand the services of certainprofessions or companies;members of MutualityAssociations; and certaincategories of state workersand administrative bodies

Members of the Casas doPovo and Casas dosPescadores and certainindependent workers couldjoin the Trade UnionWelfare Funds

166 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

the services of certain professions or companies. They covered the risks of sickness, invalidity and oldage, as well as family benefit in the case of the insured person’s death.

The third category consisted of the mutuality associations, in which membership was optional and themenu of covered risks varied according to the insurance. They were financed by members’ contributionsonly. The fourth category included the social protection institutions of state services and administrativebodies. Membership was optional for certain categories and compulsory for others and, similarly, coverageagainst risks varied depending on the institutions involved.

Despite these measures, the system and its operation were still quite unsatisfactory and inefficient.Among the main problems were the paltry pensions for farm workers and fishermen, who had nounemployment coverage either. The system provided no protection for the staff of services not included inthe institutions compelled. In addition, there was no cover for industrial accident and occupational diseaseand there was no protection for self-employed at all. No institution provided maternity benefits.

An evaluation carried out by the Câmara Corporativa11 between 1959 and 1961 demonstrated theunsatisfactory operation and dimension of the system. By the end of 1959 the total number of beneficiariesfor the Trade Union Welfare Funds and the Retirement or Welfare Funds was 863,700, plus 833,500 familymembers. Approximately 30 per cent of the workers in trade, industry and services had no cover and 40 percent of their family members fell outside the social welfare programmes. The twenty-eight Casas dePescadores12 had 54,700 registered members and 4,300 contributing members (boat-owners) but the onlyprotection the sector provided was in the form of assistance. The cover ratio for farm workers was no morethan a fifth—555 Casas do Povo (people’s funds) provided protection to 19.6 per cent of the relevantpopulation and their family members. Welfare benefits were mostly directed at well-organized sectors, suchas industry, commerce, and services, neglecting agriculture, the largest sector in terms of employment(Maia 1985). Social protection expenditure represented 2.8 per cent of GDP. Pensions represented 1.1 percent, family benefits and other benefits 1.5 per cent (Carreira 1996: 101).13

The government’s responsibilities did not include running or financing any component of a publicwelfare system, but merely the setting-up and co-ordination of institutions to attain social policy goals. Theaim was simply to create the legal framework of the system in which these institutions were to operate.More than two decades were to follow before the first reform of the system took place.

After the Second World War, when modern European welfare states began to take shape in anindustrialized Europe, Portugal was not among them. The country continued to be governed by areactionary dictator ship, conservative as regards a welfare state and officially supportive of family-basedwelfare. Industrial relations were forbidden, with a corporative state being instituted instead. Moreover, thegeneral policy was against industrialization (restricted until the 1960s) and against the modernization ofeconomic and social structures.

The 1960s were years of change in Portugal. They saw the first phase of the establishment of heavy industryand the rise of the first large industrial and financial groups. In Africa the wars of independence started. ThePortuguese population was involved in mass immigration to other countries in Europe. Social policy saw anumber of changes as well (Serrão 2000).

In 1960, in social protection institutions associated with corporative bodies, members of a family ofcovered workers gained the right to assistance in acquiring medicines. In 1960 and 1961 minimum disabilityand old-age pensions were introduced. In the latter year it was decided that family allowance for dependentminors would not be forfeited when the beneficiary retired.

In 1962 a new law14 gave the government new powers for the coordination of objectives, by putting intopractice “a joint plan for social security and other social policy sectors, in particular health and care”.Furthermore, it was given the power of decision over the activities of corporative organizations. A Social

LUÍS CAPUCHA ET AL. 167

Council of Ministers, chaired by the Prime Minister, was created. On the organizational front, the generalregulations ruling the Caixas Sindicais de Previdência15 established, for the first time, a regional (district-based) organization of the system. On the other hand, the Federação de Caixas de Previdência e Abono deFamília16 wascreated, along with the Fundo Nacional de Pensões,17 aninstitution set up in 1965 toguarantee that deferred disability, old age, death and family allowances were granted on a national scale tobeneficiaries and their families.18 The system was mainly oriented towards sickness, maternity and familybenefits, disability, old age and death pensions, covering wage earners and their families in industrial andservice sectors. Protection against involuntary unemployment and professional diseases were clearlysecondary objectives (Maia 1985). Agricultural workers saw their situation improve in the late 1960s. Thefamily benefits of this category of workers were raised to levels closer to those of the other workers. Inaddition, Law 2144/69 and Law 445/70 clarified the risks covered in the welfare regimes for agriculturalwage earners, which included sickness (medical assistance and medicines), maternity benefits, marriagebenefit, death benefit, invalidity and old age benefits. A transitory pension regime addressed to members ofCasas do Povo was created for the situations of old age and disability. Until the Law 2115 of 1962, thefinancing of the whole system was based on the capitalization of welfare funds. The new law started amixed system, allocating most of the money from contributions to pay-as-you-go pensions, reducing thecapitalization of funds, and liberating significant amounts of money to increase sickness and maternityallowances. The pay-as-you-go system became predominant, though a fund of reserve has been kept. An“assistance fund” has been created, financed by receipts other than ordinary contributions, and aimed atpaying extraordinary lump-sum benefits to beneficiaries and their relatives. The four categories ofinstitutions and the risks covered did not change. For special risks, the Caixas de Seguros19 were set up.

Apart from the changes in the agricultural workers schemes already mentioned, new schemes foroccupational diseases, sickness benefit, death benefit, assistance for sick or handicapped minor childrenallowance, and specific regulations for tuberculosis and involuntary unemployment came into being. From1965 onwards injuries and accidents at work were subject to compulsory insurance contracted withinsurance companies and financed by employers. Members of the Casas do Povo and Casas dos Pescadoresand certain independent workers were now allowed to join the Trade Union Welfare Funds.

These reforms did not change the level of the benefits, but they did allow a redefinition of thebeneficiaries, including self-employed workers, and a small extension of the risks covered. Despite theimprovements and the reorganization carried out, there were still a number of failings, in particular: thesecondary and undeveloped nature of unemployment and industrial accident protection; the multiplicity ofspecial schemes with limited reach; the low financial value of the benefits; and the financing system thatwas exclusively supported by the employees’ and employers’ contributions.

Above all, levels of cover remained very restricted. By 1970, the system covered no more than 60 per centof the active population. The number of pensioners was 187,300 and the average annual pension was stillextremely low. The minimum pension was about � 1.8 per month (Serrão 2000). Nevertheless, between 1960and 1970 expenditure on pensions rose by 17.8 per cent. In 1973 a new attempt20 at reform was made, butno significant results were achieved either in terms of population or risk coverage, or even in the value ofthe benefits.

Apart from a few marginal initiatives targeted at the homeless, who were persecuted and put ininstitutions of repression (the “Mitras”) or charity institutions (“Soup for the Poor”), which were mainlymanaged by institutions closely related to the Catholic Church, the fight against poverty remained an issueconspicuous by its absence.

Nevertheless, if we are to understand the anti-poverty policies created in Portugal after the country joinedthe European Economic Community, it is necessary to mention a community development pilot study that

168 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

took place in two small communities in the interior of the country between 1962 and 1965 (MSA 1965).This was a private initiative of a team of researchers from a Portuguese university who were trying tointroduce the Community Development Methodology into Portugal. Involving a multidimensional approach(local economic development, health, social action and infrastructure) and the co-operation of local andnational authorities, the initiative introduced the notions of participation, partnership and action-researchinto social work in Portugal for the first time. The initiative was obviously limited in its direct impact, sinceno more than two small communities were involved, but it did play an important role in the training of anew generation of social workers who did their practical training there. It was this generation that took overthe leadership of most of the projects and initiatives developed after 1986, under the II European Anti-Poverty Programme and the National Programme to Combat Poverty. They were indeed the professionalsand decision-makers who were responsible for supporting and launching these programmes.

After 25 April 1974: towards universal protection

The pro-democracy military coup on 25 April 1974 was motivated by the discontent that prevailed among asector of the armed forces and the most politicized pro-democracy sectors in the society. The roots of thedissatisfaction lay in the colonial war and the under-development that was holding the country back. Therewas a sense that the regime had become unsustainable and its end was in sight. When the end arrived, therepercussions were felt at all levels of the country’s social, political and economic fabric. On the one hand,the goals of political democratization and economic and social modernization, which in Portugal largelymeant Europeanization, were supported by the military reformers and the parties of the Socialist andChristian-democratic block (the PS and the PPD/PSD) (Guillén and Silva 2001). These were confronted, onthe other hand, by the demands of the movements influenced by the Communist Party and the militarypersonnel whom this party controlled within the Armed Forces Movement, in a revolutionary process aimedat installing a Soviet-style socialist regime in Portugal. This revolutionary period lasted until 25 November1975, when the stabilization process of the democratic government was set in motion, culminating in the1976 Constitution of the Republic (Fields 1975; Harsgor 1976; Harvey 1978; Graham and Wheeler 1983;Kayman 1987; Clijsters 1992).

These particular circumstances brought a rapid expansion in social policies and social protection, with theestablishment of a set of measures, which had a great impact on Portuguese society and the State. Apartfrom the far-reaching changes affecting the liberties and guarantees of social and labour rights, it was at thistime that the national minimum wage was introduced; pensions had their minimum value doubled and theirmaximum value restricted; the non-contributory regime for the social security system was set up, whichallowed the introduction of the social pension for anyone over 65 years of age or any disabled person over14, for whom the system had until then merely provided attendance allowances and for whom there was nocover in any welfare scheme; unemployment benefit, medical assistance and medicine supplies for bothsickness and maternity and family allowance for unemployed contributors’ family members were extendedto all dependent workers and their families; agricultural wage-earners who were not members of the Casasdo Povo,21 were covered by the welfare funds of those institutions and new benefits were created for themand their families. The Health and Social Welfare Services were also set up. Moreover, there was animprovement in the administration process to access benefits, namely a reduction in the waiting periodbetween the application and the date of payment.

These new measures, combined with others in the fields of labour, health and education, provoked asharp rise in public expenditure but also produced the broad effect of redistributing income and providinguniversal social protection. This led to the consideration that a welfare state had been formally set up in

LUÍS CAPUCHA ET AL. 169

Portugal, still with important gaps in protection, in particular in terms of safety nets, but a far cry from theunder-developed Bismarckian system that had hitherto existed. This situation was immediately reflected inthe system’s main performance indicators.

In 1975, 78 per cent of the working population was covered by the different social protection schemes asactive contributors. The number of pensioners rose from 187,300 in 1970 to 861, 700 in 1975. Between1970 and 1975 pensioners numbers in the rural regime increased from 22,000 to 490,000 and in the generalregime pensioners numbers increased from 43,000 to about 371,000.

On the financial front, social solidarity (i.e. non-contributory benefits) became a national responsibility.Social protection expenditure rose from 2.8 per cent of GDP in 1960 to 7.5 per cent in 1975 (with pensionsrepresenting 3.7 per cent) (Carreira 1996:101).22 Pension expenditure as a percentage of total socialexpenditure increased from 14.9 per cent (1960) to 45 per cent (1975), at an average annual growth rateabove 36 per cent. From 1965 health services increased the number of beneficiaries by 2.6 million to almost7.3 million in 1975. Accordingly, more than 86 per cent of the population then had access to health care.

These achievements, however lasting in terms of their design, registered a drastic slowdown in the periodof democratic consolidation after the political crisis of 25 November 1975, when the Communist Partyproject was defeated. The revolutionary period was marked by great instability, in the context of extremelyadverse economic and financial conditions due to the world oil crisis, capital outflows, serious socialconflict, low productivity and chaos within the productive system, among other factors. The Socialist andChristian-democratic bloc, which held power between 1976 and 1986, the year Portugal joined the EEC,defined the preparation of membership as its top priority, on the basis of a broad national consensus. Thismeant that they not only had to stabilize and consolidate democratic institutions but also had to adoptextremely tough and austere macroeconomic measures, which led to a marked retrenchment in the nation’sliving conditions, an increase in unemployment and erosion of the purchasing power of wages andpensions.

This was no obstacle to progress in social protection policies. One important step was the creation of theNational Health System in 1979.23 Social protection was also a field of deep changes. A significant part ofthat progress was related to the organization of the system. The Instituto de Gestão Financeira da SegurançaSocial24 was set up, the Centros Distritais de Segurança Social25 were established and supervision and debtcollection services were strengthened. Specific schemes were improved. Paid maternity leave was extendedto 90 days. A social protection scheme was set up for independent workers, covering sickness (medicalassistance and the supply of medicines), maternity, disability, old age and death. Domestic workers wereincluded in the General Welfare Scheme. This period of designing and consolidating the social protectionsystem, culminated in the Lei de Bases da Segurança Social of 1984, which remained in force until 2000.This law established a system organizing the existing schemes under the headings “contributory”, “non-contributory” and “social action”, as illustrated in Table 5.2.

The new law brought no new expansion to the system.26 Indeed, this measure fits into the logic of auniversal service and a rise in benefits levels which characterized Portuguese social protection policies fromEEC membership in 1986 until 1995 (Guillén and Silva 2001).

At this stage Portugal took advantage of a positive change in the international economic situation arisingfrom the rebound from the oil crisis in 1986, economic growth, the devaluation of the dollar, and,consequently, the fall in the costs associated with the external debt. It also benefited from the CommunityCohesion Funds, which had a significant impact on the movement of capital and, at the same time, allowedthe state to carry out a number of infrastructure initiatives connected with communications, health,education, occupational training, human resources development programmes and technological conversion

170 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

support programmes for business. This helped to sustain rapid economic growth, which only suffered aslowdown—or a reverse—in 1993, and a significant increase in social protection expenditure.

In 1980 social protection expenditure in Portugal was only 11 per cent of GDP (against an average of 22.2 per cent within the other member states of what is now the EU), though it rose to 16.3 per cent in 1990and then to 23.9 per cent in 2001. Social expenditure per capita (PPS) rose from around 26.8 per cent of theEuropean average in 1980 and 39.1 per cent in 1990 to 51.9 per cent in 1995 and 56.9 per cent in 2001.27

This increase mirrored the convergence of the GDP (which stood at 55.2 per cent of the European averagein 1980 and increased to 60.7 per cent in 1990 and 70.6 per cent in 1995), though it was even more markedthan that of the latter, which reflects the considerable priority given to the

Table 5.2 Lei de Bases da Segurança Social (1984)

Contributory regime Non-contributory regime Social action

Target population Dependent and independentworkers and their relatives

Families, the disabled andthe aged when not coveredby the contributory regime

The needy Families anddependants Poor urban andrural neighbourhoods

Components General compulsory schemeAlso RSSV (VoluntarySocial Security Scheme),RESSAA (Special SocialSecurity Scheme forAgricultural Activities) andautonomous socialprotection system for civilservants

Non-contributory socialprotection schemes

Social assistance Householdservices and socialequipment NationalProgramme to CombatPoverty

Main risks and schemes Sickness, maternity,occupational hazard,unemployment, disability,old age, death and familybenefits

Child support through familybenefit Maternity andsurviving dependants’benefits Social pension anddisability benefit

Economic and socialbreakdown Socialdiscrimination Social andcommunity developmentDependent family members’needs (long-term care)

social area. Half the distance to the European average in terms of the intensity of social expenditure hadbeen covered.

The first consistent anti-poverty policies also date from this time. Indeed, in spite of the continuingimprovement in social policies, there had been no consistent set of policies to combat poverty in Portugal.28

These arrived with the II European Anti-Poverty Programme in 1986 (Commission of the EuropeanCommunities 1991), a European initiative that was in place when Portugal joined the EC. This programmewas based on the promotion of action-research projects oriented towards target groups like the long-termunemployed, young unemployed, elderly, single-parent families, migrants and refugees, drug abusers,people developing criminal and anti-social behaviour and poor neighbourhoods in urban and rural areas.The approach, putting great emphasis on promoting participation and partnership between different publicand private actors, was supported and managed by social action services led at the time by some of thesocial workers that had been trained in the pilot project of the 1960s mentioned above. The results werewidely recognized and given great visibility, providing room for the launching of the National ProgrammeAgainst Poverty (PNLCP),29 supported by the national social action budget. Later, the local projectssupported by the programme were designed on the principles of the Third European Anti-PovertyProgramme, calling for the participation of target groups in the projects, multi-dimensional approaches

LUÍS CAPUCHA ET AL. 171

(including housing, health, education, training, employment, income distribution and other areas),partnership, involving local authorities, national government and private bodies, and territorially-basedinitiatives. These initiatives were to address the actual factors determining poverty in the areas, which,finally, included a sizeable part of the country. Until the second half of the 1990s, the PNLCP was the mainsystematic approach to poverty in Portuguese policies.

The social situation: recent trends

The social context—some data

Economic growth was particularly marked in Portugal following the entry to the European Community.From 1986 to 2001 GDP grew by an average of 3.4 per cent per annum, although it has declined since2002. Similarly, productivity growth declined from 1.8 per cent in 2000 to 0.2 per cent in 2002, as happenedto a similar extent in the rest of Europe (0.4 per cent). Portugal’s productivity rate is still the lowest,representing 68.4 per cent of the EU average. This is explained, on the one hand, by the relativebackwardness of the economy (low average levels of human resources skills, a sectored structure withserious imbalances, competitiveness based on low labour costs and a work organization that is neither verymodern nor generally innovation-friendly). On the other hand, there are high levels of employment in therelatively unskilled and unproductive sectors.

Nevertheless, the country’s economic performance brought about an improvement in the livingconditions of the population. It was supported by a set of effective distributive policies, particularly in thelast five years of the 1990s, though significant problems still remain in this respect.

Income distribution and poverty

Regardless of average improvements of living conditions, poverty remains widespread in Portugal. The fewmajor quantitative studies carried out in the country point in this direction. Ferreira (2001) using data fromthe Household Budget Surveys of 1980/81 and 1989/90, sets a poverty line based in basic needs, andestimates that the population falling below this line increased from 24.1 per cent to 26.3 per cent during thisperiod (see Table 5.3).30

Table 5.3 Poverty rates in Portugal by relevant features, 1980/81 and 1989/90

1980/81 1989/90

Poverty rate Sample (%) Poverty rate Sample (%)

Poverty rate—by age

<35 13.7 15.0 16.8 13.2

35–44 21.4 18.5 23.2 18.7

45–54 22.6 21.6 20.8 19.6

55–64 21.7 20.7 25.0 21.4

65–74 32.0 16.8 32.6 17.2

75 or more 45.6 7.4 48.1 9.9

Poverty rate—by sex

Male 21.9 80.1 24.1 79.4

172 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

1980/81 1989/90

Poverty rate Sample (%) Poverty rate Sample (%)

Female 33.2 19.9 34.9 20.6

Poverty rate—by household type

Single person (<65) 22.2 4.0 – –

Single person (>65) 46.0 6.8 – –

Single person – – 35.8 11.8

2 adults 21.3 25.7 26.4 27.0

3 or more adults 15.8 16.4 19.0 15.3

Single parent with dependent children 29.3 2.0 34.3 3.0

2 or more adults with dependent children – – 25.7 42.9

2 adults with dependent children 21.0 26.1 – –

3 or more adults with dependent children 31.7 19.0 – –

Poverty rate—by most frequent activity status

Independent farmer 24.7 13.5 37.3 6.7

Farm worker 34.7 5.5 38.5 3.4

Employers (excl. farmers) 15.1 9.2 21.5 11.8

Managerial staff 1.8 4.3 3.9 5.3

Industrial employees 19.0 21.3 32.4 20.1

Commercial employees 9.0 14.6 17.8 14.3

Inactive 39.4 25.6 26.4 33.5

Other economically inactive 33.1 6.1 39.3 5.0

Poverty rate—by income source

Employment 16.1 48.6 17.1 48.3

Self-employment 23.2 26.0 21.6 17.8

Income on capital 19.2 1.9 8.3 3.0

Pensions and benefits 47.9 19.1 51.4 25.1

Other sources 17.7 4.4 18.7 5.8

Poverty rate—by level of education

No education 36.5 40.5 46.8 26.8

Basic education 18.8 48.2 22.9 57.7

Secondary education 2:8 8.9 4.8 12.2

Higher education 1.0 2.5 0.6 3.3

1980/81 1989/90

Poverty rate Sample (%) Poverty rate Sample (%)

Poverty rate—by region

North 29.8 33.6 29.8 33.1

Centre 20.6 21.6 29.6 19.3

Lisbon and Tagus Valley 17.1 33.1 20.4 33.4

LUÍS CAPUCHA ET AL. 173

1980/81 1989/90

Poverty rate Sample (%) Poverty rate Sample (%)

Alentejo 39.2 8.2 33.7 7.5

Algarve 23.6 3.5 20.7 6.6

Poverty rate—by territory

Rural 29.8 56.4 31.7 54.9

Urban 16.8 43.6 19.7 45.2

Poverty rate 24.1 100.0 26.3 100.0

Source: Adapted from Ferreira (2001).

Most of the poor in Portugal are working poor and old age pensioners. Farm workers are also heavilyaffected by poverty situations both in 1980 and 1990, and the risk increased throughout the decade. Thiswas also registered for independent farmers and industrial employees. Old people are much more affectedby poverty than the younger generation, and that matches with high poverty rates of people with noeducation, as does poverty among farm workers. Women, single adults (mainly older adults), single parentfamilies and extensive families are to be found among the most vulnerable categories. In 1990 only theregions of Lisbon and Tagus Valley and Algarve registered a smaller proportion of the population inpoverty relative to the national average. In the 1980s poverty was higher in rural than in urban areas.

According to another study using the same micro-data (Rodrigues 1993) and according to our owncalculations,31 the poverty rate (defined as 60 per cent of the median equivalized income of individuals;OECD equivalence scale) declined slightly from 17.7 per cent in 1980/81 to 16.8 per cent in 1989/90,increasing again to 17.7 per cent in 1994/95. During the first decade of 1990 the poverty gap32 grew from 0.22 per cent to 0.25 per cent.

In an analysis of the distribution of average disposable income per adult equivalent, the ratio between thelast decile and the first decile (i.e. the income of the richest 10 per cent and of the poorest 10 per cent)increased from 7.3 in 1990 to 9.2 in 1995 (in 1980 it was 7.8). The ratio between the last vintile (the richest5 per cent) and the first vintile grew even more significantly, going from 10.9 in 1990 to 14.5 in 1995. In thesame period, the Gini coefficient went from 0.32 to 0.35, which reflects greater inequality in the distributionof income.

Poverty in Portugal, understood either as the absence or scarcity of resources of all kinds needed tosatisfy basic needs (food, housing, education, work, training, security, health, protection, access to cultureand leisure, among others) or as relative deprivation (in comparison with lifestyles that are considerednormal), is much more related to historical development processes than to exclusion arising from brokenties with institutions or social networks. The phenomenon is widespread but its outlines are defined, in mostcases, by the persistence of marks left by both under-development in the economy and by the shortcomingsof social protection (Capucha 2000) ,33

Nonetheless, the rapid modernization processes in the country in recent decades have brought verysubstantial improvements in living conditions for most of the population but, as in the rest of theCommunity, they were not able to prevent the emergence of cases of “new poverty”. Single-parent families,mainly those whose adult is a woman with little or no access to childcare services, are one example of thesesituations. Drug abusers, HIV/AIDS victims, the homeless, children at risk, prison inmates and ex-convictsare other “modern” categories in a particularly problematic situation of social exclusion. For quite differentreasons, and requiring an equally different political framework, two other categories at risk of exclusion

174 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

have resulted from the recent dynamics of Portuguese society: immigrants and very elderly dependants34

(DEPP/MTS 2002:215–252).In the second half of the 1990s, the first results of the European Community Household Panel (ECHP)

were published and this became the source of information used in all official documents on poverty, bothregarding European processes and national poverty-reduction policies. The main indicators for measuringpoverty and inequality in the distribution of income were now also those adopted at a European level.35

According to data from the 2001 wave, poverty36 affected 20 per cent of the Portuguese population, whilethe European average was 15 per cent. Persistent poverty also remained high, as 15 per cent of thepopulation had lived below the relative poverty line for the current year and at least two of the precedingthree years, as opposed to 9 per cent of the population in the EU as a whole. Both poverty situations haddecreased during the second half of the 1990s. As shown in Table 5.4, income inequality also showed adecrease in the same period. This slight drop occurred in the context of a steep rise in average income, andthe poverty threshold per single-person household went from � 2,602 in 1995 to � 3,589 in 2000 (a 30.6 percent rise, similar to that for a household of two adults with two children). All indicators point to the samedownward trend in the poverty indicators and the continuing existence of a considerable difference from theEU average.

As shown in Table 5.5, single-person households are particularly vulnerable to poverty, with a povertyrate of 39 per cent (after social transfers). Women (43 per cent) and the elderly (46 per cent) are thecategories that are affected most in this type of household. Indeed, the elderly are more

Table 5.4 Poverty and income distribution in Portugal (%)

Main indicators 1995 1996 1997 1998 1999 2000 2001

Poverty rate (in %) 23 21 22 21 21 21 20

before social transfers (pensions excluded from socialtransfers)

27 27 27 27 27 27 24

persistent (current year and at least 2 out of preceding 3years)

– – 15 14 14 14 15

Gini coefficient 37 36 36 37 36 36 37

Share ratio (S80/S20)—equivalized income 7.4 6.7 6.7 6.8 6.4 6.4 6.5

At risk of poverty threshold for a single person household (in� )

2,602 2,788 2,967 3,017 3,168 3,397 3,589

At risk of poverty for a household with two adults and twochildren (in � )

5,464 5,856 6,232 6,336 6,652 7,134 7,538

Source: Euros tat, Newcronos.Note–=not available.

Table 5.5 At-risk-of-poverty rate by household type in Portugal (%)

Poverty rate-by household type 1995 1996 1997 1998 1999 2000 2001

Single person 48 48 45 45 45 42 39

Single male 44 43 37 38 35 38 28

Single female 50 49 48 48 49 44 43

Single person (<30) – – 16a 13a 0a 2a 1a

single person (30–64) 34 35 30 33 34 36 28

LUÍS CAPUCHA ET AL. 175

Poverty rate-by household type 1995 1996 1997 1998 1999 2000 2001

Single person(>65) 57 55 53 52 52 47 46

Single parent with dependent children 34 34 44 45 39 37 39

2 adults (both <65) 21 18 19 18 12 15 13

2 adults (at least one =65 or over) 41 38 41 40 37 38 32

2 adults, one dependent child 13 15 13 11 13 11 9

2 adults, two dependent children 16 15 15 12 16 18 15

2 adults, 3 or more dependent children 45 37 46 50 39 36 49

3 or more adults 15 14 15 11 11 9 10

3 or more adults with dependent children 22 19 19 19 22 24 23

Poverty rate 23 21 22 21 21 21 20

Source: Euros tat, Newcronos.Notes–=not available,a Unreliable.

vulnerable than average, even when two of them live together, as the poverty rate is 32 per cent among two-person households in which at least one of them is aged over 65. Households with two adults and three or moredependent children also show an abnormally high poverty rate (49 per cent). Larger families are, in fact, morevulnerable than average, as the poverty rate among households with three or more adults and dependentchildren is 23 per cent. Indeed, these categories that of classic families of two adults with two children andthat of single-parent families were the only ones whose poverty rate went up between 1995 and 2001. Thelowest poverty rates are found among households consisting of three or more adults and no children,couples with only one child and childless couples.

Employees are the socio-professional category least vulnerable to poverty. The poverty rate amongemployees is 7 per cent and in 2001 it was 14 per cent for people whose income derived from work. It isimportant to note, however, that, in Portugal, having a job and living from one’s salary cannot be consideredas a guarantee against poverty, given the figures for these categories. This idea gains even more weightwhen we consider the fact that the highest poverty rate (28 per cent) is among self-employed. The povertyrate for the unemployed is 38 per cent, while for retired and other inactive people the rate is as high as 28per cent (Table 5.6).

The persistent nature of poverty, as well as its profile, seems to indicate

Table 5.6 At-risk-of-poverty rate by activity status and income source in Portugal (%)

1995 1996 1997 1998 1999 2000 2001

Poverty rate—by most frequent activity status

Employed—but not self-employed 8 7 8 8 8 8 7

Self-employed 35 33 31 30 28 30 28

Unemployed 30 28 31 32 34 23 38

Retired 34 31 32 30 29 28 25

Other economically inactive 27 26 26 23 23 28 28

Poverty rate—by main income source

Income from work 16 15 15 15 13 15 14

176 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

1995 1996 1997 1998 1999 2000 2001

Private sources 32 22 28 43 25 19 26a

Social transfers 49 46 46 43 47 40 43

Poverty rate 23 21 22 21 21 21 20

Source: Ferreira (2003) and Eurostat, Newcronos.NotesIncome from work—wages and salary earnings, and self-employment income.Private income—capital income, property income, and private transfers received.Social transfers—unemployment benefits, pensions, other social transfers.–=not available.a Unreliable.

that it cannot be explained by short-term situations or by ruptures in people’s lives. On the contrary, tounderstand it we must look at the way the systems operate, i.e. the labour market, demographic and familytrends and the social protection system.

The labour market and education

Participation in the labour market is one of the main mechanisms for social inclusion. This does not implythat a direct relationship can be established between employment/unemployment levels and poverty, sincemost people and families in a situation of poverty are or once were (in the case of retired people) fullyintegrated into the labour market.

Portuguese employment indicators have been among the most favourable in Europe. Portugal had in 2002an activity rate of 72.1 per cent, while the EU average was 69.7 per cent. The Portuguese employment ratewas 68.1 per cent compared to 64.3 per cent in the EU. The unemployment rate was 5.1 per cent, while theEU rate was 7.7 per cent.37

Employment showed sustained growth between 1995 and 2001, going from 63.8 per cent to 68.6 percent, in spite of the drop between 2001 and 2002 (from 68.6 per cent to 68.1 per cent). The growth ofwomen’s employment was more marked, starting at an already high level (54.4 per cent in 1995), and goingup to 60.8 per cent in 2002, one of the highest in Europe. The percentage of women in the employedpopulation rose from 44.4 per cent in 1998 to 45.1 per cent in 2001. This aspect of the Portuguese labourmarket, which contrasts with the other southern European countries, had its roots in the 1960s. At the time,the female employment rate was only 13.1 per cent (Almeida 2000). The beginning of the colonial war in1961 and the exodus of emigrants in the 1960s, in which more than 1.5 million Portuguese went abroad,resulted in women joining the labour market in great numbers. The changes in the employment structure inthe last decades favoured women’s continued presence in the market, with the growth of services (mainlyhealth, education and administration, on one hand, and personal and domestic services on the other), whichaccounted for 64.6 per cent of employed women in 2002 (jobs in services correspond to 53.8 per cent oftotal employment). Moreover, the manufacturing industry specialized in traditional industries using femalelabour, such as clothing and textiles, and employs 21.7 per cent of women. They are also employed inagriculture though to a lesser extent (13.7 per cent), although this percentage is slightly higher than the totalweight of the sector, which is 12.4 per cent.38

The unemployed population decreased from 1996 to 2000, reaching a minimum of 205,600 (3.9 per cent)(Table 5.7). In 2001, the number of unemployed increased slightly to 215,600 (93,200 males and 122, 500

LUÍS CAPUCHA ET AL. 177

females), and the rate went up to 4.1 per cent. Between 2001 and 2002, unemployment accompanied theeconomic cycle and went up by 1 point

Table 5.7 Main employment indicators

1998 1999 2000 2001 2002

Totalemployment (15to 64 years)

4,863.3 4,928.7 5,028.9 5,098.4 5,106.5

Male 2,703.7 2,717.0 2,767.7 2,799.7 2,796.2

Female 2,159.6 2,211.8 2,261.2 2,298.7 2,310.3

Totalemployment rate(%)a

66.8 67.4 68.3 68.6 68.1

Male 75.6 75.6 76.4 76.5 75.7

Female 58.3 59.5 60.5 61.0 60.8

Totalunemployment

253.7 226.8 205.6 215.6 272.3

Male 111.6 108.9 89.4 93.2 122.0

Female 192.1 117.9 116.2 122.5 150.3

Totalemployment rate(%)

5.0 4.4 3.9 4.1 5.1

Male 4.0 3.9 3.1 3.2 4.2

Female 6.2 5.1 4.9 5.1 6.1

Total long-termunemploymentrate (12 monthsand more)b

115.5 93.7 89.7 85.9 101.2

Male 49.8 43.3 40.8 36.9 43.7

Female 65.7 50.5 49.4 49.1 57.5

Source: INE (Inquérito ão Emprego).Notesa Employed population (15–64 years)/Total population (15–64 years)×100.b From 2001 the total (“unemployment duration spell”) does not correspond to total unemployed population. It does not

include individuals who will start working later in the future (within the next 3 months).

to 5.1 per cent. The gap between the male and female unemployment rates was high and even grew between1995 (1.7 percentage points) and 2002 (1.9 percentage points). The average gap between male and femaleunemployment in the European Union was not very different from that in Portugal in 2002, standing at 1.8percentage points.

Despite the positive trend in labour market indicators, structural problems still persisted. First, eventhough the annual long-term unemployment rate was only 1.5 per cent in 2001, the weight of long-term intotal unemployment was about 40 per cent, making the situation of those involved particularly problematic.On the other hand, the percentage of households in which no-one was employed in the total population agedfrom 0 to 65 was only 5.4 per cent in 2002.39

178 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Second, there is the low level of schooling of the active population. In recent years, the most qualifiedpopulation has played a major role in employment growth, and important changes have occurred,particularly among the younger population. The 2001 census attests to an increase in the population’seducational achievement level when compared to 1991. This, in turn, reflects a decrease in the population withless than secondary education (67.4 per cent to 57.1 per cent), and substantial growth in the population withupper secondary education (8.7 per cent to 15.4 per cent). The proportion of the population with highereducation rose from 4.9 to 10.6 per cent. However, compared to the situation in other EU countries, thedifference remains high. According to Eurostat,40 the population aged between 18 and 24 with less thansecondary education and not at school nor in training is currently 45.5 per cent, which is higher than in1995. The poor level of participation by low-skilled workers in vocational training underlines this weaknessin the Portuguese labour market. In fact, the proportion of the population aged 25–64 attending educationand training in the four weeks prior to the survey for the Eurostat data was only 2.9 per cent in 2002 (8.5per cent in the EU). The situation seems to have worsened, since the proportion was between 3.3 per centand 3.5 per cent from 1995 to 2001. Curiously, in this particular case, as indeed in all the indicators onparticipation in the education system, the situation of women is a little better, with a life-long learningindicator of 3.3 per cent in 2002, as opposed to 2.4 per cent for men.

Third, low-skilled work is related to a pattern of economic specialization that relies on low labour costs.This pattern can be found in a large segment of companies focusing mainly on traditional industries such asshoe-making, clothing, textiles and furniture, and on construction, retail trade, low-skilled services andagriculture. Work tends to be organized in a traditional manner and the quality of work is often poor, indetriment to workers and employers in terms of security and flexibility. This, of course, includes an absenceof services and work arrangements to help women reconcile their work and family lives. Relatively lowwages are the main factor of competitiveness for the low-production segment companies. Thus, in 1998,productivity per worker in Portugal was no more than 66.2 per cent of the European average,41 workers’average pay was only 67.5 per cent of EU-15, and the proportion of workers on low salaries (those makingless than two-thirds of median earnings) was 11.0 per cent in 2000, in spite of the decrease from 14.1 percent in 1995. The pressure to keep salaries low has been reinforced by the existence of an extensive blackeconomy.42

Family and ageing

According to provisional data from the last Portuguese census (2001), the number of households has beenrising. Large and extended families have been decreasing, as families with five or more people, whichrepresented 19.8 per cent of the total in 1991, corresponded to 11.4 per cent in 2001. Conversely, smallfamilies have increased significantly. The number of one-person households, which increased about 45 percent between 1991 and 2001, is well above the variation rate of total families.

The emergence of new family models, with increasingly individualized units, is associated with profoundchanges in the forms of solidarity that, for a long time, represented an alternative to inadequate publicwelfare. Women who have no access to personal and domestic services (the majority in fact) are especiallypenalized in a double sense, since they still have to do the household work, while also holding down a job.43

The elderly population living alone is another vulnerable group. In 1998 there were close to 593,000elderly people over 75, and between 1991 and 1998 the number of dependent elderly people—of whom 78per cent were women—grew by 18.6 per cent. The marked ageing of the Portuguese population is one ofthe most relevant features of recent demographic developments. In the 2001 census, the elderly (65 yearsold or more—16.4 per cent) outnumbered the young (0–14 years old—16 per cent) for the first time. This

LUÍS CAPUCHA ET AL. 179

phenomenon highlights specific problems that social action policies and household services will have toaddress. In the specific case of childcare, the role of grandparents in child minding is growing lessimportant (Torres et al. 2000). It is more common for working mothers to seek care outside the family.Where the elderly are concerned, there are situations in which the family provides all the care necessary,but most of the time this is no longer the case. Since only a minority can afford to hire direct services ontheir pensions, a social answer is often the only solution.

These trends, along with the emergence of such phenomena as drug abuse and HIV/AIDS and with achange in demands for the rights of people with disabilities, stress the need for a good network of socialfacilities. Figure 5.1 shows the development of social facilities by grouping.

Children and young people are one of the priority areas. Existing childcare services are already full tocapacity. The actual coverage rate for the population under 3 years old is, however, only 19.8 per cent.Concerning the older population, 93.1 per cent of the capacity is being used in elderly care services, and theactual coverage rate is 11.6 per cent for people over 65. Coverage rates have proved to be insufficient tocope with basic needs. The informal market typical of household services is one of the most commonalternatives used to respond to unsatisfied needs.

Sodal protection: recent trends

As shown above, since 1975, Portugal has made a significant effort to catch up with its Europeancounterparts in terms of its level of expenditure on social protection (see Table 5.8), regarded as the mainmirror of a country’s social effort. In 1980, expenditure on social protection represented 12.8 per cent of theGDP (24.3 per cent in Europe). It rose to 17.2 per cent in 1991 (EU average 26.4 per cent) and to 22.7 percent in 2000 (27.3 per cent in the EU). This shows that the gap was drastically reduced, though there is stilla significant difference.

After Portugal joined the European Economic Community in 1986, there was a period of “massificationwith differentiation” of social protection (Guillén and de Silva 2001). The system is now entering itsmaturity phase. There were a total of 1,518,689 old-age pensioners in 2000, accounting for 60.9 per cent ofall pensioners and for 36.5 per cent of all social protection beneficiaries. Disability pensions covered 373,

Figure 5.1 Evolution of social facilitiesa by target (source: DEPP/MTS, Carta Social).

Note

a Infrastructures for delivery of care, such as childcare or elderly care.

180 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

337 people (15 per cent of all pensioners) and 602,277 people were receiving a survival pension (24.1 percent of pensioners). Pensioners represent 25 per cent of the Portuguese population. The total number ofdirect recipients in the whole social security system increased from 4,197,818 in 1995 to 4,276,826 in 2000.

The amount of the benefits has been increasing at a fast rate—the minimum pension under thecontributory system rose 24.3 per cent, the social pension under the non-contributory system increased 42per cent, and the average pension grew by 37 per cent between 1995 and 2000. Still, in terms of adequacy,benefit levels remain relatively low: a minimum pension of � 189.50 in the general system and a socialpension of � 138.30 in the non-contributory system (corresponding to 35.5 per cent of the minimum wage).

According to Eurostat (European Communities 2001), in 1999 social exclusion expenditure isapproximately 1.7 per cent of total social protection benefits in Portugal (1.6 per cent in the EU). In additionto the effort these figures illustrate, other recent measures must be taken into account to give an overallpicture of the impact of the system as a whole on poverty alleviation.

Table 5.8 Social protection expenditure (% of GDP)

1991 1996 2000

Portugal 17.2 21.2 22.7

EU-15 26.4 28.4 27.3

Source: Eurostat, Newcronos.

Between 1995 and 2000 there was a period of development of the system. It was characterized by areform known as the “new generation” of active social policies introduced by the Socialist Party governmentthat won the elections in 1995. The main lines of this reform were:

• The assertion of the State’s prime public responsibility to promote protection policies and, especially, tofund measures aimed at national solidarity with the most disadvantaged groups and to ensure thesustainability of the system in terms of inter-generational solidarity and equity in labour solidarity.

• The introduction of the principle of “positive discrimination in social protection measures”, namely inthe pension system and family allowances, to make them more flexible.

• Taking up the fight against poverty, whose existence was officially recognized, as one of the priorities ofnational policies.

• The reorganization of the public social security system to bring it closer to its beneficiaries andcontributors, to improve nationwide co-ordination against appropriation by local interests and to clarifythe responsibilities of three subsystems (the social welfare and citizenship subsystem, the familyprotection subsystem, and the welfare subsystem).

In 2000, a new revision of the Social Security Law44 was implemented. While maintaining the three corecomponents, the new law aimed at both improving the level of the benefits, and ensuring the sustainabilityof the system. There was a change in the criteria used to calculate allowances in the pension system—theformula now includes the person’s whole contribution career.45 The public pension trust fund wasreinforced. The financing of social assistance and social solidarity measures was set up as the exclusiveresponsibility of the state budget. The reforms, which are still ongoing, include the introduction ofinstitutional organization measures, answering to the need for proximity between services and beneficiaries.An agreement supporting this reform has been signed by four of the five social partners.

LUÍS CAPUCHA ET AL. 181

The new policies for social inclusion

The promotion of inclusion46 in Portugal has been one of the government’s priorities since 1995, aimedspecifically at reinforcing the dynamics of inclusion in Portuguese society (DEPP/MTS 2002). Theconstruction of an innovative strategy has been achieved in several different ways. Along withdevelopments in social protection benefits, major changes have been made in such areas as care for theelderly, childcare, programmes to fight poverty and social exclusion, training, employment, housing, etc.Some of these are summarized below.

The National Programme against Poverty

The National Programme against Poverty is an instrument for local action, based on the experience gainedduring the European anti-poverty programmes Poverty II and III, and is intended to achieve integrateddevelopment through a series of local projects. The basic idea of the programme is that the main problems ofpoverty in Portugal are related to development and that community-based action is an appropriate tool infighting it.

One of the most marked situations in Portuguese society in recent decades has been a significanttransformation in the patterns of territorial occupation. This process is fourfold. It involves thedesertification of the rural interior, urban growth, the concentration of the population along the coastline,and the emergence of two metropolitan areas in Lisbon and Porto. The rural interior is thus gradually beingeconomically and socially deprived and is showing a growing lack of investment, infrastructure andservices to ensure acceptable living conditions for its inhabitants. As a result, its population consists moreand more of poor pensioners and traditional peasants, who rely considerably on plural incomes.

On the other hand, on the outskirts and in the centres of the metropolitan areas problematic quarters areemerging in which the vulnerable categories not only tend to concentrate, but also to proliferate. In fact,they create circles of persistent poverty, where poor housing, marginalization, illegal residents, low-paidjobs, the black economy, unemployment, difficulties in access to basic infrastructure, services andequipment, and frequently low-quality institutional performance, among other forms of deprivation, tend tomean that poverty is persistent and transmitted from generation to generation. Some of the most problematicconsequences of poverty often originate or are located in these urban communities.

Two regional committees manage the National Programme against Poverty, one for the North and the otherfor the South. They are responsible for funding, monitoring, and assessing the local projects. The averagelength of a project is four years. To address the multiple dimensions of poverty in a specific area, theprojects usually include responses that range from physical subsistence to participation in community life.Action covers areas such as the creation of social facilities and services, vocational training, basic trainingand education, the development of craft industries and other forms of production, support for self-employment, local revitalization of productive activities, the reconstruction of housing and maintenance. Afew projects are targeted not on areas but on specific risk groups, such as the homeless.

These projects work as network partnerships. There is cooperation between the public and private sectorsand integrated inter-sector action. In this respect, local authorities, associations of municipal councils,and non-governmental organizations (NGOs) are encouraged to act as promoters and a council of partners iscreated for each project. In addition, the active involvement of the target population is encouraged.

The main handicap of the programme is, on the one hand, its limited capacity to produce profoundimpacts on economic local systems, and, on the other, the excessive dispersal around the country, instead offocusing resources on more problematic areas.

182 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

The Social Network Programme

The Social Network is a structural programme for local development and social policies. Its main focus isthe implementation of local strategic (municipality-based) planning processes, as a foundation for socialintervention. This methodology requires the use of reported social diagnostics, the implementation of localinformation systems and the negotiation of social development plans. It does not intervene directly insolving the problems of individuals and groups in situations of or at risk of poverty, but it calls for theintervention of all relevant actors and for the mobilization of appropriate resources. In short, the philosophyof the Social Network Programme is based on four essential principles: integration, co-ordination,innovation and subsidiarity.

The consolidation of these networks, which are intended to be effective and dynamic partnerships, isembodied by local social action councils (CLAS, at a municipal level) and in the parish social committees(CSF), platforms for the planning and co-ordination of local social intervention, capable of mobilizing alllocal citizens.47

This programme ran on an experimental basis in forty-one pilot municipalities in 2000. In 2003, there arealready 176 municipalities with social networks (from a total of 308) whose activities are in varying stagesof development, involving an average of fifty partners per local project.

The main problem of the Social Network is the fact that it is extremely dependent on the actual interestand involvement of the municipalities’ mayors. On the other hand, since it involves a reform of theorganizational principals of the “welfare mix”, namely, a new design in the relations between localgovernments and NGOs in social action at local level, the programme is somewhat vulnerable toconservative and corporative interests.

The Social Employment Market

The Social Employment Market (MSE) is a decisive instrument in providing access to economic activity forthose who, for personal, social or professional reasons, have no other way of obtaining work. It provideseither a permanent solution or, preferably, a transitional path into the open market. It consists of economicactivities supported by the State to satisfy needs that are not met by the market and an integration system forthe employment of disadvantaged groups. It includes a series of existing measures (occupationalprogrammes for the unemployed, craft school workshops, social companies and protected employment fordisabled people) and makes room for the creation of new measures, such as special vocational training.

The Social Employment Market combines two innovative principles in Portuguese employment policies:on one hand, the development of individual employability through measures that run parallel to insertioninto economic activity and, on the other, effective decentralization resulting from the involvement of a widevariety of public and private institutions in promoting the measure.

The main challenge facing this measure lies in the difficulty of its objectives. In particular, specialtraining48 is highly demanding, in terms of technical and pedagogical complexity and is time-consuming,and in terms of mobilization of resources.49 On the other hand, there has been a tendency to replace thesolutions aimed at promoting access to the open market by continuing social support. Moreover, actiontowards economic efficiency is often replaced by the rationality of traditional social action. Nevertheless,the impact of social employment market cannot be ignored, since it supports each year more than 69,000people who have no other resources for employability, training or occupation.50

LUÍS CAPUCHA ET AL. 183

Towards a universal safety net—an overview

One of the most obvious expressions of Portugal’s historical underdevelopment was the lack of a system ofsocial policies capable of providing universal protection for citizens, leaving it to families in general, andrural families in particular, which represented almost half of the population before 1974, to provideessential care in case of need. Even in the sector of insured workers, their protection was highly deficient,both in terms of the risks covered and especially in terms of the quality of the protection provided.

Three decades after the revolution, the structure of activities has changed completely and the old ruralsociety, where only in the 1960s did industry of any significance first appear, has become a society whereservices predominate (especially those linked to retail trade, tourism, financial activities, real estate, andeducation, health and protection policies) and where the secondary sector, dominated by construction and thetraditional footwear, textile and furniture industries, employs almost a third of the population. In turn, ademocratic regime was installed after the coup d’état of 25 April 1974 and the country began its“Europeanization”, i.e. it started to imitate the patterns of the other European welfare states.

These processes have not, however, been unilinear or homogeneous. Family and demographic patterns,for example, soon developed characteristics similar to those of the more advanced countries, while thequalifications of the active population still remained much lower. Economic activities were divided betweencompanies that modernized, in terms of technology, management practices, qualifications of humanresources, organization of work and competitiveness, and invested in innovation, and companies, in fact themajority, oriented to maintaining low labour costs. As a result, the country’s employment rates are some ofthe highest in Europe, including female employment, but the quality of work, average productivity andwages in substantial segments of the economic tissue are far below the average European levels.

A brief appraisal of the incidence of these contrasting social and economic dynamics in poverty shows,on one hand, a significant reduction in the phenomenon, which was particularly marked in the twenty yearsfollowing the revolution, in the wake of economic growth and the overall improvement in the population’sliving conditions. On the other hand, Portugal still has one of the highest poverty rates in the EuropeanUnion. Only a small part of this poverty is caused by social exclusion or the “breaking” of social ties (withwork or family) resulting from the transition to the new economic, social and cultural scenario of theknowledge economy (Rodrigues 2000), the risk society (Beck 1992), or the individualization of socialintegration mechanisms (Giddens 1991). Most of it is fuelled by the generational reproduction of circles ofpoverty and ingrained marginality, by poorly paid workers and by elderly pensioners.

The weight of elderly pensioners in the poor population, of which they are the largest category, is themost notorious example of the kind of challenges to the effectiveness of the welfare state, which is facedwith a double mission. On one hand, it has, with some delay, to complete the construction of policiestypical of the Fordist industrialization period, upgrading the quality of social protection system, education,health and other social policies. On the other, it has to reform social policies and their design, in order torespond to the new phenomena of ageing, unemployment, precarious employment, social exclusion and theemerging realities of the new economy.

The first part of this mission basically began immediately after the coup of 25 April 1974 and thetroubled times that followed it. In addition to the restoration of civil and political liberties and thereplacement of the corporatist regime by free trade unions and employer’s associations, the universal rightto pensions was introduced and minimum non-contributory social pensions were created. Social protectionexpanded rapidly. The minimum salary was introduced and all salaries went up. The universal right to State-provided health care and medical services was introduced, among other measures that, by law, make up awelfare state. They were, however, created at a time of serious economic recession and of disinvestment,

184 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

exodus of capital and great conflict in political and labour relations. A legal and institutional basis had thusbeen built but it lacked the resources to ensure proper funding and operation.

A large number of these advances occurred under a revolutionary impetus strongly influenced from thepro-Soviet Portuguese Communist Party. It was opposed by a front of the Socialist and Christian DemocratParties, sometimes officially in opposition and informally converging in action, and at other timesconstituting a formal coalition. The Socialists won the first free elections in 1975, followed by the ChristianDemocrats, and the political stabilization process began in 1976. One of its priorities was to re-establisheconomic and financial balance in the hope of joining the European Communities, which became a nationalgoal. In spite of the constraints imposed by two interventions from the International Monetary Fund,51 thefoundations were laid at the time for the national health system and the social security system. The socialsecurity system was organized around contributory and non-contributory schemes and social action. Thecontributory scheme covered the main risks of sickness, invalidity, old age, occupational accidents,unemployment, death, widowhood, orphan-age, and motherhood. It also included family benefits forfamilies with children, students or disabled members. The non-contributory scheme provided, as well asfamily benefits, minimum coverage for the risks of old age, invalidity and death. Social action involved anincipient network of social facilities to support families and their dependants and also provided smallamounts of aid, depending on available funds in incidental situations of risk or extreme need.

As already discussed, in 1986 Portugal joined the European Economic Community, and this marked aturning point for the country. Benefiting from the devaluation of the dollar and the resulting reduction in theexpense with public debt, from the favourable economic situation, and from the revitalization of themovement of capital stimulated by the common market and the arrival of the first structural funds, theeconomy grew fast for six years. With the support of the structural funds, the Social Democrat government(centre-right) first elected in 1985 and afterwards legitimized by absolute majorities in two consecutiveelections in the late 1980s and early 1990s, gave priority to the construction of infrastructure in the areas ofcommunications, health, education and sanitation. Social policies were also expanded and reinforced,however. A national vocational training and employment support system was created from the existingemployment services. Access to higher education was generalized. In the field of pensions, as we have seen,this was a period of growth of the benefits, improvement in the coverage of the population and an increasein the system’s general performance. Portugal joined the II and III European Anti-Poverty Programmes,whose projects, supported by a number of social workers who had trained as part of the pilot project in 1965(see p. 212) and by now held positions of responsibility in the social action system, stimulated the launch ofthe National Programme Against Poverty.

The crisis of 1993 brought with it a number of problems and high-lighted others which had beenconcealed by the expectations generated by the accentuated growth of the preceding years. Unemploymentgrew to levels close to those of the democratic and economic stabilization period of 1976/86. Povertyincreased, after having gone down for twenty years, and inequalities grew. Particularly problematiccategories emerged or gained visibility, such as the homeless, the long-term unemployed and single-parentfamilies.

In this depressing context, the Socialist Party won the elections in 1995 with a programme that promisedto reinforce policies aimed at developing human resources and made the fight against poverty one of itsmain priorities. The guaranteed minimum income (GMI) was debated during the electoral campaign andwas to become the banner of the socalled new generation of social policies, through which Portugal wastrying to catch up with its European partners in the agenda to reform their social model.

LUÍS CAPUCHA ET AL. 185

The guaranteed minimum income (GMI)

In 1996, Law 19-A/96 introduced a guaranteed minimum income scheme in Portugal, aimed at the poorestindividuals in society. The GMI constitutes a minimum baseline income to cover essential needs in order tomaintain a basic level of human dignity (and is thus independent of the national minimum wage, whichcovers the minimum rates employers must pay their employees). According to the assessment report on theexperimental stage, this represented a dual movement which, on one hand, led excluded people, familiesand groups to begin processes favouring access to rights of citizenship and social participation and, on theother, led institutions to offer real opportunities of starting such processes, providing them with means andsupport (Capucha et al. 1998).

The Portuguese scheme is based on the idea of combating social exclusion. According to the scheme, theemphasis on job-seeking is combined with measures in different areas, such as vocational training,education, health, access to household services or housing, and other areas where self-sufficiency can beachieved. Second, the onus of combating the causes that bring people to the scheme is not unilaterally borneby the applicant, but results from an agreement between the individual and the community. This stressessociety’s joint responsibility in the reintegration process. According to this innovative principle, GMI isbasically a contract providing an income (a benefit from the non-contributory system, recognizing acitizen’s right to a minimum standard of living) in exchange for a commitment to participate in a socialintegration programme.

It is also a subsidiary measure to other social benefits, which means that beneficiaries first have to claimthose other benefits. In fact, GMI may take the form of a benefit supplementing other social benefits. Inthese cases, the scheme will cover the difference between the amount of the benefit and the GMI.

GMI was the result of political recognition of the need for effective responses capable of directlysupporting the income of vulnerable individuals and/or families, promoting their occupational and socialqualification, stimulating and providing them with the opportunity for direct participation in society andgiving them the notion of citizenship.

The socio-economic context of GMFs birth was marked by a recent increase in inequality, whichoccurred during the crisis at the beginning of the decade. It had interrupted a cycle of strong economic growth,which was then resumed. Poverty was widespread (approximately 23 per cent), and extremely harsh insome cases. On the other hand, social protection systems were experiencing difficulties resulting from thefact that they had not yet reached maturity. The state was committed to rigorous macroeconomic measures,which were aimed at economic integration and the single currency. Considerable political innovation wasthus necessary.

The guaranteed minimum income in Portugal was the subject of political debate during the parliamentaryelection campaign in 1995. The European Council recommendation, made during the Portuguese presidencyin 1992, suggesting that all member states should introduce some sort of basic income policy, was thenquoted. Poverty and social exclusion gained visibility, reinforced by the approval and launch of the schemein 1996.

Having become one of the “flagships” of the new government, with a prominent role in the context of the“active social policies”, GMI benefited initially from wide consensus in society and a vast base of socialsupport. The value of solidarity, public awareness of the phenomenon of poverty, and the fact that itincluded a contractual insertion clause all contributed towards this consensus.

It was initially introduced for one year on an experimental basis. The experimental phase was to helpdecide on some aspects of regulation, namely to make it possible to define and organize the local structures,to adjust the procedures for verifying incomes and defining the configuration of the households and the

186 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

members to be included, to try out different forms of implementation and contents of the insertion contract,to determine details of the resources and income to be considered in the benefit formula.

Initially, progress was made through a small number of scattered local projects. The number wasgradually increased, and the experimental phase became a transition to universal measure. The experimentalphase made it possible not only to clarify aspects of regulation, but also to hold a national debate on themeasure and its objectives. The first experimental stage thus tested the feasibility and consistency of themeasure so that it could be accompanied by the gradual strengthening of the structures and resourcesavailable to make it more effective in the future. The guaranteed minimum income came into force on 1stJuly 1997.

The design of the scheme

Eligibility requirements

The Portuguese legal framework governing GMI lays down a range of selection criteria (Table 5.9) basedon personal characteristics (nationality, residence, and age), resources and willingness to work. GMI appliesto all citizens in a situation of proven acute economic need, regardless of nationality, who are legallyresidents in Portugal. The latter condition (legal residence in the country) could, to some extent, become abarrier for the homeless in which case an institution (for example, a NGO) can be designated as a legalresidence.

Age is an additional requirement. Claimants must be at least 18 years old with some exceptions: pregnantwomen, young people with dependent children and dependent people with disabilities.

Furthermore, the amount awarded varies according to the resources/ income available to the claimant,taking into consideration the income earned by the other members of the household.

All the beneficiaries (all members of the household) must be actively available for social andoccupational (re)integration, which will be applied by signing up for a social integration programme.

Benefit calulation

GMI offers a variable allowance. The applicant’s household income is assessed and the allowance to begranted corresponds to the difference between a person’s existing income and the GMI. Thus, an importantstep

Table 5.9 Main aspects of GMFs original design

Selection criteria Definition

Duration 12 months, extended automatically

Nationality Regardless of nationality

Residence Legal residence in Portugal (the homeless give the address of a NGO)

Age At least 18 years of age, or under 18, if the person is emancipated and has dependentchildren, or in the case of pregnancy

Willingness to work Availability for employment (condition suitable for those whose age and state of healthallow them to work), as well as training and other activities aimed at social integration

Exhaustion of other claims Applicants and all members of the household must have exhausted all other benefits towhich they may be entitled

LUÍS CAPUCHA ET AL. 187

in the process of implementing the scheme is the measurement of net income, determining the exact amountto be paid. The consideration of the overall income of the applicant’s household includes, in addition to theapplicant’s own income, the income of his/her spouse or partner and of other people depending on theapplicant’s resources, particularly his/her relatives.

The household includes the spouse or a person living with the beneficiary in a de facto relationship formore than one year, under-age relatives (and also over 18 in the case of students and people not required tobe available for employment), adopted children, and children under the guardianship of the applicant.

In Portugal, the amount of the benefit is tied to the amount of the social old-age pension (which isupdated annually on the basis of the consumer price index), which for 2003 is � 124.70 per month.52 Thesize of the household is taken into account in calculating the amount of the benefit. The proportion to becovered by the GMI therefore corresponds to the difference between the total income of the recipient’shousehold, irrespective of its nature and source, and the total amount of the benefit to which s/he would beentitled, given the composition of the household. The following equivalence scale applies:

• 100 per cent of the old-age pension for the first two adults;• 70 per cent of the old-age pension for each adult from the third;• 50 per cent of the old-age pension for each minor.

Family allowances, student grants and 20 per cent of net income from training grants and work (after alldeductions) are not included in the calculation of the household’s total income. The same applies to 50 percent of wages and grants from training in the first year for those beneficiaries who join the labour market.53

The amount of the benefit may be supplemented by other supplementary benefits (potential allowances forextra support to make insertion agreements feasible, which are therefore awarded on the basis of the area inquestion, e.g. education, health, or transport costs for people attending evening classes). The amount ofthese benefits varies according to the need, and the maximum annual amount per household is 24 times thesocial pension. Whenever the GMI allowance falls below 5 per cent of the minimum income (95 per centsocial pension GM), this figure is the amount to be granted.

The GMI is paid monthly through the local offices of the Regional Social Security Centres. Thebeneficiaries are eligible for one year with the possibility of automatic renewal, provided that thebeneficiary continues to meet all the requirements and criteria listed above. The forms submitted byclaimants are checked when the claim is made. For purposes of means testing, claimants must submit all thenecessary documentation attesting to their economic situation. If they are unable to submit docu ments, aformal declaration will be accepted. Social workers from local social action offices must assess the currentsituation of the claimant/ beneficiary. This reassessment is part of a means test procedure and may lead toadjustments to or even the end of the benefit, if the household’s current income and declared income arevisibly different. All the partners in the Comissões Locais de Acompanhamento (CLA)54 of the GMI canparticipate in this assessment, thus monitoring the fairness of the declarations.

When the household’s income is subject to contributions, the means testing procedure consists ofchecking in the Salary Register at the district centres. When no income is declared, a written statement fromthe interested party is accepted and, if there are any doubts or if any of the CLA members are aware of ahigher income, home visits are made to check the actual situation.

The average amount of the benefits in December 2002 was � 150.1 per household and � 51.3 per individual.Total expenditure on benefits reached its peak of � 299,278,000 in 2001, and decreased to � 239,422,000 in2002 (Table 5.10).

188 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Social integration contract

A social worker is in charge of each claimant, following the individual approach and should produce areport, which is included in the household’s file. The report must include a description of the problems thatthe household and its individuals face and should propose an integration plan.55 The approval andmonitoring of integration plans are the responsibility of the CLA set up at municipal level, comprisingrepresentatives of regional bodies working in social security, employment, vocational training, educationand health and other public and private bodies. The social integration plan and its administration costs arefunded by the State budget, in line with the Social Security Framework Law.

One innovative aspect of the scheme is the inclusion of a contract in the integration programme. It takesthe form of an agreement signed by

Table 5.10 GMI expenditure

GMI expenditure (103 � ) GMI expenditure in % total social expenditure

1996 1,760.8 0.0094

1997 48,657.7 0.2421

1998 197,194.8 0.8906

1999 2,773,96.5 1.1370

2000 2,667,97.0 1.0040

2001 2,442,81.3 0.8305

Source: INE, Contas Nadonais; IGFSS, Conta da Segurança Social; Eurostat, Newcronos.

the social worker appointed by the CLA, the beneficiary and other members of the household. The processtherefore requires close joint co-ordination between the CLA, which appoints a social worker in charge ofmonitoring and assessment, and the beneficiary, with a view to preparing and implementing a personalizedprogramme in line with the needs and resources of both parties. The range of obligations to be fulfilled bythe beneficiaries and the support given by the institutions at local level are thus defined. The integrationagreement must also include reference to all the support to be granted, the entities responsible formonitoring and the obligations accepted by each of the parties involved.

The integration agreement may be renegotiated with the beneficiary if it is unsuitable or needs to beamended. The whole process is legally binding, and one of the conditions allowing the individual to gainaccess to the benefit is that s/he must be willing to start an integration plan.

The GMI represents a new approach to methods of administering social assistance through the applicationof a new philosophy of action, which consolidates the ideas of guaranteed rights, negotiated integration anddecentralized and participative management. The innovation lies in the idea of a contract and in local co-ordination.

Organizational features

The creation of the GMI was accompanied by the establishment of the Comissão Nacional do RendimentoMínimo (National Minimum Income Commission) at the central level, and of the CLA. The NationalMinimum Income Commission involves representatives of the Ministries of Education, Health, Labour andWelfare, and Justice, the Secretary of State for Housing and representatives of local authorities, privatewelfare institutions and social partners.

LUÍS CAPUCHA ET AL. 189

The National Minimum Income Commission thus acts as a link between several ministries and sectors.According to the law, the duties of the National Commission incorporate three tasks:

1 Assessment: Assessing legislation on minimum income and of the social effectiveness of the measure;preparation of the annual report on the implementation of the minimum income scheme.

2 Mobilization: Monitoring and supporting the activities of the bodies responsible for applying the GMIlaw.

3 Innovation: Submitting proposals to amend the legal framework with a view to improving it andadapting it to the real social situation.

The National Commission receives technical support from the IDS (Institute for Social Development),56

which also manages the GMI.One of IDS’ responsibilities was to provide administrative and technical support for the CNRM,

monitoring the application of the GMI law, elaborating the annual report on the implementation of theminimum income scheme, and preparing draft amendments to the legal framework of the scheme. The IDSholds regular meetings that are attended by the regional GMI coordinators, mainly to provide informationon the main difficulties facing implementation.

At local level, the GMI is based on the local monitoring committees (CLA). They are responsible formanaging the measure.57 The CLA are local structures for coordination between partners working at thelocal level. Each CLA is responsible for a municipality. When a municipality is over-populated or there ismarked geographic dispersion that justifies the establishment of another CLA in the same territorial area,the competent Regional Social Security Office may put this into effect in coordination with the localauthorities.

The CLA must consist of:

• a representative of the local social action services, on behalf of the Ministry of Labour and Social Affairs;• a representative of the local employment services, also on behalf of the Ministry of Labour and Social

Affairs;• the municipal adult education co-ordinator, on behalf of the Ministry of Education;• a representative of the health centre on behalf of the Ministry of Health.

The law also provides for a range of institutions to sit on the CLA, provided that they state their availabilityand willingness to do so, and pursue their activities in the geographical area of the CLA. These bodies are:

• local governments;• private welfare institutions;• employers’ associations and trade unions;• other legally established non-profit-making bodi dies.

A member of the staff of the Regional Social Security Office co-ordinates most of these CLA, although thelaw allows other bodies to play this role, provided the majority of the partners agree.

The responsibilities of the local monitoring committees fall into three distinct fields:

1 Operation: Creating sections when justified; appointing the institutions (besides the Regional SocialSecurity Office services) responsible for receiving minimum income benefit claims; appointing the

190 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

institutions that can be chosen as the legal residence of claimants who do not have a permanent abode;and appointing the Executive Board.

2 Co-ordination of resources: Analyzing needs and capacities in the area with a view to developingintegration programmes; promoting integration activities to be adopted locally; and liaising with otherCLA and with the National Minimum Income Commission.

3 Planning and assessment: Approving annual action plans and periodically assessing the degree ofcoverage and execution of the integration programmes under way and their effectiveness.

The CLA operate in plenary sessions. They can, and usually do, have an Executive Board consisting of fouror six members in addition to the co-ordinator. Where the co-ordinator of the CLA is not a representative ofSocial Security System the latter must be a member of the Board.

The Executive Board meets regularly. The CLA continue to draw up plans of action and annualevaluation reports. Local action plans are systematically processed by the National Minimum IncomeCommission, which provides the CLA with the information needed for their own planning. The CLAevaluation reports are systematized and published in an annual performance report.

Generally speaking, some difficulties are also experienced in diagnosisbased planning with clearlydefined objectives, goals and evaluation indicators. The measure’s monitoring team provides specialsupport charts to facilitate this planning.

Beneficiaries: a profile

Numbers of beneficiaries

From the start of the measure until December 2002, 826,974 people were involved (about 8 per cent of thetotal population), of whom 320,155 (around 3.1 per cent of the population in 2002) were receiving the GMIbenefit at the end of 2002 (Figure 5.2).

Given the tight means test criteria, of a total of 479,657 applications evaluated, 194,099 (40.5 per cent)were rejected, The reason for this was twofold: 79.3 per cent were due to income levels above the minimum,and 13.7 per cent did not submit any evidence of income (tax return, personal declaration or otherdocument). GMI beneficiaries are more numerous in the North (35.6 per cent) and Lisbon and Tagus Valley(24.5 per cent), which are the regions with the greatest concentration of poor individuals58 and households,and also the most populated areas.

After a growth period between 1997 and 1999, the scheme seems to have reached maturity and thenumber of beneficiaries continues to decrease (see Figure 5.3). A significant number of beneficiaries arecurrently considered “reintegrated into society” after a period of assistance from the minimum incomescheme. By December 2002, 175,979 households, corresponding to 506,819 individuals, had stoppedreceiving this monetary benefit. Around 62.7 per cent of these people left the Minimum Income schemebecause they were no longer in economic need (see figure 5.4).59

This situation is associated with the insertion programmes developed.60 By December 2002, 75,379integration programmes had been established, corresponding to 229,853 social integration actions andcovering 173,257 individuals. The integration programmes were distributed among different areas. Socialservices (such as childcare facilities, psychosocial

Figure 5.4 Reasons for withdrawal from GMI (source: Instituto para o Desenvolvimento Social, December 2002).

LUÍS CAPUCHA ET AL. 191

assistance, family education and counselling, and support for specific groups of the population—womenvictims of violence and abuse, the homeless, drug addicts, immigrants, etc.) involve the greatest number ofbeneficiaries, representing 34 per cent of the total number. This is followed by health (23.6 per cent),education (14.8 per cent), and employment (15.7 per cent). Housing and vocational training have lowerpercentages (see Figure 5.5).

Taking into account Portuguese structural problems, it is worth mentioning the total of 13,850beneficiaries who are attending ongoing, recurrent (or “second chance”) education, and the 16,871 childrenand young people who have currently returned to school.

Figure 5.2 Annual evolution of GMI beneficiaries over the total resident population (%) (source: Instituto para oDesenvolvimento Social, December 2002).

Figure 5.3 GMI annual beneficiaries and household representatives (source: Instituto para o Desenvolvimento Social,December 2002).

192 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Most employment insertion measures take place in the social employment market. We might expect thepercentage of beneficiaries in need of a

Figure 5.5 Insertion areas (%) (source: Instituto para o Desenvolvimento Social, December 2002).

job to be higher, but many of them were already employed (about 25 per cent of beneficiaries), thoughearning a very low salary. On the other hand, about a half of them do not meet legal criteria foroccupational insertion. Anyway, a set of measures was launched in order to improve the access toemployment by GMI beneficiaries. The following are some examples.

Horizontes 2000 is a programme aimed at accelerating the employment insertion process by using amethodology based on a joint interview between the social worker responsible for the household, theemployment services, and the beneficiary, aiming at the immediate preparation of a draft personal employmentplan adapted to the individual’s situation. This programme may include the following active employmentmeasures: direct placement in labour market, self-employment initiatives, social enterprises, and theProgram Insertion-Employment. The programme can offer the following benefits in health, education andhousing. Health: child vaccination, commitment to a specific treatment plan and family planning, treatmentof drug addiction and alcohol; education: second-chance education courses and assuring children schoolattendance; and housing: support for remodelling work and purchase of domestic appliances.

Characterization of the beneficiaries

The majority of the beneficiaries are women, representing 69 per cent of both applicants and beneficiaries.Another prominent feature of the structure of the GMI beneficiary population is the coverage of the youngerage groups. 39.9 per cent of the beneficiaries (127,511 people) are under the age of 18, many more than the27,512 beneficiaries aged between 19 and 24 (see Figure 5.6). Adults aged from 25 to 64 years oldrepresent 43 per cent of the beneficiaries, while people over 65 represent 8.4 per cent.61 The group aged 19to 24 represents 8.6 per cent. Concerning the situation of GMI beneficiaries in the labour market, 24.2 per

Figure 5.6 Beneficiaries by age group (%) (source: Instituto para o Desenvolvimento Social, December 2002).

LUÍS CAPUCHA ET AL. 193

cent are employed, 23.8 per cent are students, 19.8 per cent are unemployed, and 14.7 per cent are at home(see Figure 5.7). Only 9.1 per cent are pensioners, while 4.5 per cent are unable to work.

The underlying family structure of the GMI beneficiaries corresponds to the most traditional type offamily—couples with children. Other types of family carry significant weight in this population, particularlysingle-parent households (mostly mothers), who are the second main GMI beneficiaries (see Figure 5.8).

Implementation problems and recent debates

The implementation of a scheme with the above-mentioned characteristics is usually beset with a series ofdifficulties to which the GMI is cer tainly not immune. The first notable public criticism appeared in late1999 following a report by a team of experts (though none in social policy, or more broadly in socialsciences) from the Audit Department of the Exchequer. The report highlighted the less positive results ofthe measure and identified some irregular payments, although empirical data showed little evidence of the

Figure 5.7 Situation of the beneficiaries of GMI in the labour market (source: Impact evaluation of the GMI, December2000–March 2001, IDSCNRM).

Figure 5.8 GMI types of households (%) (source: Instituto para o Desenvolvimento Social, December 2002).

194 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

fact. Subsequently, the commitment to combat fraud was reinforced, while attention was drawn to the factthat GMI fraud was quite insignificant when compared to tax evasion and to fraud in other social measures.The number of rejected claims showed clearly that the control mechanisms were working. At the time, afraud inspection campaign was launched in all social policies, and showed that the inappropriate award ofthe subsidy under the GMI scheme was no more than 11.6 per cent, against, for instance, 29 per cent in thecase of sickness benefit.

In spite of its low technical quality, the report from the Audit Department of the Exchequer had apolitical relevant effect. The Popular Party took the opportunity to make a “right-wing” critique of thescheme and to bring a number of issues into debate again. Most of these issues focus on questions such asdisincentive to work and dependence on the benefit (poverty trap), thus questioning the real efficacy of theintegration objectives.62 The topic of fraudulent access to the measure emerges in this context. Two otheraspects have also been criticized from “inside” the scheme. The first concerns the shortage of humanresources (qualified professionals) for individual approaches (namely, means testing) and to accompany theinsertion process. The other focuses on partnerships, arguing that the State maintains an authoritativetradition in its contact with beneficiaries and that institutions are not capable of offering adequate, high-quality insertion contracts. There has also been a lively debate on the costs of the scheme, which could becontrolled either by reducing or freezing benefit levels, or by creating more restrictive conditions forentitlement.

The government implemented the Horizontes 2000 programme mentioned above and companies wereinvited to fill in the requests of employment through the services, to improve the adjustment between supplyand demand. The outcomes of these measures were limited, although this created a filter to prevent a fewbeneficiaries who worked in parallel and informal activities, from remaining in the system.

The main reason for disincentives in the scheme is, indeed, to be found in the value of the minimumincome itself. In principle, there are only conditions that may lead to the “poverty trap” in the case of part-time or odd-job workers with extremely low wages. Furthermore, many of the beneficiaries are employedworkers or their relatives. On the contrary, GMI has permitted the activation of many people who had beenexcluded from the labour market for generations.

Related to fraud, there exists another problem of a very different nature. The legal framework that set upthe GMI lists the different sources of monetary household income that must be taken into considerationwhen assessing and quantifying its resources. The framework for assessing non-monetary income is,however, not clearly established in the law. In Portugal non-monetary income represents roughly 13.5 percent of net household income (MSST 2003) and tends to be important for some poor categories of thepopulation such as, for instance, peasants and gypsies, whose case was then quoted. Failure to take non-monetary income into consideration, though not frequent, may cause distortions, which are often used in thepropaganda against the measure.

The other main criticism has to do with the effectiveness of the integration contracts. Some institutionsthink that the results achieved are good. Others say that the evidence of households that do not leave themeasure proves its inefficacy. There is no benchmark to allow for objective evaluation. But a problemremains above this controversy. The human resources are insufficient for close, personal monitoring of theintegration processes. Although the GMI has brought about a significant increase in the number of socialworkers recruited by the Regional Social Security Offices, the idea still remains that the resources continueto fall short of demand. Operating in partnership by establishing the CLA was expected to allow the humanresources of the various partners to be fully exploited in terms of sharing responsibilities. But theparticipation of other partners is expressed in domains other than the provision of staff. As a result, socialworkers often have difficulty in covering all the households under their responsibility.

LUÍS CAPUCHA ET AL. 195

Certainly, the problem is not only about the numbers, but also about organization of work and resources.The Regional Social Security Offices have not yet adapted to new work methods that could fully implementmanagement by objectives, defining priorities, teamwork and fighting bureaucracy.

In spite of the initiatives undertaken to facilitate the insertion processes, like Horizontes 2000, the FAINA(Support Fund for Insertion in New Activities), the social empl loyment loyment loyment ported byEuropean Social Fund Operational Programmes, social workers sometimes feel dissatisfied with the resultsof training and socio-professional integration. The argument is that the amount of resources allocated is notsufficient. The existing programmes do not always take into account the necessary tools and methodologies(including time) to provide beneficiaries with adequate action, from a multidimensional perspective. Thisresults from an asymmetry in the system. In fact, the latter is designed to depend upon the capacity of localpartners to find the ways—and to innovate in their practices—to provide real, good-quality opportunities forsocial insertion of the beneficiaries. But while these can be punished if they do not play their part in theagreement, there is no sanction if institutions do not provide good opportunities. Therefore, thebeneficiaries’ responsibility is over-stressed.

Though not as prominent as the ones previously mentioned, one final difficulty has to do with non-take-up. Several indicators show that lack of information is not a real problem. However, they also show that thecomplexity of the application procedures, and the associated social stigma, both deserve consideration. Thefirst and only existing study focusing on this issue indicated, on the basis of the analysis of the HouseholdBudget Survey, that 4.8 per cent of individuals and 5.7 per cent of households should be entitled to thesubsidy (Rodrigues 2001). As seen above, these figures were achieved, even exceeded. Hence, there is littleempirical evidence of situations of non-take-up.

The campaign for the last parliamentary elections (March 2002) stressed the need to cut publicexpenditure and to reduce public debt. The poverty trap argument and the costs of GMI were used again bythe Popular Party (that became a partner in the government’s coalition) to argue in favour of changes in thedesign of the scheme. It must be noted that none of the electoral programmes, even of parties that had beenagainst it in previous elections, mentioned eliminating the GMI.

The new right-wing government (formed by the Social Democratic Party and the Popular Party) made therevision of the design of GMI one of its first measures. The changes introduced by the new law andregulations, however, were minimal. Apart from the name (GMI became RSI— Social Insertion Income),they were basically reduced to the need of beneficiaries to present new applications every year; theconsideration of income in the past twelve months for benefit calculation; a small reduction of the ceiling forcomplementary benefits; introduction of payment in kind for expenses with health, housing and access tohousehold equipment and facilities; reinforcement of the complementary benefits for pregnant women andmaternity and also for households with disabled people; and finally new rules for control of fraud.Organizational changes also occurred, but not specifically in the GMI scheme: the IDS was integrated intothe new ISSS (Institute for Social Security and Solidarity), keeping the functions and basic structures.

Since the de facto changes were so small, it can be said that the minimum income scheme has beenaccepted by all the political parties and is now fully integrated in the Portuguese welfare system. Thedebates will probably move, therefore, from the scheme’s design and legitimacy, to the question of efficacy.This will shift the attention to the questions of organization at local level and human resources allocated tothe new RSI, in the context of restrictive budgetary policies in order to control the state deficit, thegovernment’s priority, which constitutes a barrier to the reinforcement (if not a reason for shortage) of thoseresources.

196 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Conclusion

In the past three decades, the State in Portugal has undergone a series of changes that have taken it from theunder-development typical of an authoritarian state on the “semi-periphery” of Europe (Santos 1990) to awelfare state of unfinished modernity (Viegas and Costa 2000), belonging to the European Union and, likeits partners, involved in an open co ordination process aimed at social inclusion.

GMI has played a major role in recent developments in the field of social policies. With GMI, the Statecommitted itself to adding the principle of national solidarity with the most disadvantaged to the traditionalreciprocity between contributors to the social protection system based on the principle of inter-generationalsolidarity. Indeed, this is in line with a broader type of reform that permeated the whole European socialmodel at the time. GMI introduced the principle of equity in the relationship between the State and socialassistance beneficiaries by contractualizing measures aimed at activating people who thus become jointlyresponsible in the process of their own dignification and social inclusion.

The principle of the personalized and contractual approach to citizens was later taken even further in theinitiatives designed to respond to the European Employment Strategy Guidelines, with early activation ofthe unemployed, building a personal plan for reinsertion into the labour market. Almost at the same time,the principle of positive discrimination63 was being tried out in family benefits, where before there was onlyuniversal equalization. This principle is, furthermore, being extended to other areas.

The principle of partnership between the different public, social and civil actors, which is embedded inthe design, management, execution and assessment of the GMI, institutionalized the idea of welfare-mixpolicies. They bring the system as close as possible to problems and allow better focusing of publicintervention on the actual needs of each family and each person, as subjects and not merely the objects oftheir relationship with the institutions. This principle was in the basis, for instance, of the vast expansion ofthe social employment market, creating the paths for transforming the beneficiaries into participants insocial life. The creation of the Social Network Programme generalized the principle of the welfare mix as amodus operandi, while the reinforcement of the National Programme against Poverty illustrated thedissemination of the principle of territorially based intervention.

Social policy developments created the climate that favoured social dialogue. In early 2001, medium-termagreements were reached with the social partners on “employment, training and education”, “occupationalsafety and hygiene”, and the “reform of the social security system to ensure its sustainability in the future”.This reform reinforced the principles of public responsibility, of inter-generational solidarity insustaining pensions in the future, of national solidarity in the fight against poverty (social action measureswere fully absorbed by the State Budget), of proportionality between contributions and benefits, reflected inthe change in the period counted when calculating pensions, which was extended to the person’s wholeworking life. The institutional organization of the system was also subject to reform, creating the Institutoda Solidariedade e Segurança Social (Institute of Solidarity and Social Protection) and the Lojas daSolidariedade (Social Protection front offices) with the double aim of bringing the system closer to citizensand taxpayers and improving national coordination.

In the context of economic growth that characterized the second half of the 1990s, the new generation ofpolicies contributed to a 3 percentage points reduction in poverty and also to a positive impact oninequality.

When it took over the presidency of the European Union in 2000, Portugal tried to push further anti-poverty policies at the European level. Portugal suggested the addition of a third axis, social cohesion, to the“employment and economic growth” that had been first proposed as the subject of an annual spring meetingof the European Council. The European social inclusion process based on the open method of co-ordinationwas then ready to be launched at the Nice Summit (Larsson 2002; Ferrera et al. 2002).

LUÍS CAPUCHA ET AL. 197

The first National Action Plan for Inclusion (NAPincl), by which Portugal, like the other member states,responded to the four objectives in the fight against poverty agreed on at Nice for the period 2001–03, hadtwo main aims. The first was to harmonize the large number of old and new measures for fighting povertywith other policies directly or indirectly aimed at social cohesion in the area of labour, lifelong learning,housing, health, etc. It was a question of imbuing the existing instruments with political unity, emphasizingthe transition from the assistance model to the active, contractualized promotion of inclusion. The secondaim was to expand some of the principles of the new generation of active social policies, extendingexperiences and introducing innovative practices that had already proved to work.

The NAPincl 2001–03 established the goals of reducing relative poverty and absolute poverty (thoughwithout clearly defining the last concept), and eradicating child poverty. The plan had three axes forresponding to these objectives, taking into account the nature of poverty and social exclusion in Portugal.The first axis involved the development and modernization of existing universal systems, such as workorganization, qualification of human resources, development of a comprehensive lifelong learning strategy,increased productivity and the future sustainability of the pension system, to mention only the most decisiveareas. There was the conviction that a good part of the poverty problems in Portugal involved economic andsocial development.

The second axis was the prevention of poverty and concentrated on taking pre-emptive measures againstunemployment, increasing the lowest pensions, making the redistributive function of the social protectionpolicies more effective, creating an observatory for household indebtedness and expanding the network ofservices supporting families and the conciliation of work and family life.

The third was designed to deal with the most serious cases of poverty and exclusion mobilizing existinginstruments, such as the social labour market, special training, the social network, the National Programmeagainst Poverty and the Guaranteed Minimum Income. New instruments were created in the meantime, suchas the Social Emergency Network articulated with a “national social emergency line”, to deal with people atrisk of serious social rupture, such as victims of violence, children at risk, the homeless and other equallyurgent situations. It provides immediate support and refers them to systematic, professional assistance.

Other measures were announced, such as new programmes to promote focused and integratedintervention in the most disadvantaged urban and rural areas (fifty “urban social development contracts” inrun-down neighbourhoods and slightly fewer municipalities involved in the new “Rural Areas and SocialDevelopment” programme). Following the GMI model, the personalized approach to each person in need orin exclusion, with a view to agree upon signing up to a social insertion programme, was also foreseen in thefirst NAPincl.

For all these reasons, the NAPincl 2001–03 became a new instrument to support the intended transitionfrom an unfinished protectionist welfare state to a mature and active one. The European process bringslegitimacy to a double objective. On the one hand, this new welfare state should concentrate on the task ofcompleting the construction of the “traditional” policies that were lagging furthest behind, in the field ofhousing, social facilities, and education, for example, and, on the other, it was meant to be involved in thefront line of active social policies, capable of combining universal protection with differentiatedintervention focused on disadvantaged groups and capable of negotiating the activation of individuals fromthese groups on the basis of the principle of social citizenship. It also aimed to improve the government’sability to develop national and local level co-operation agreements and structures between the differentgovernment departments, local authorities, and social and civil partners.

At the beginning of the new century, the economic environment became more hostile and the difficultiesof supporting the growth of the budgetary outlay on social policies increased, one of the reasons being the

198 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

euro stability pact, which imposed rigid budgetary restrictions. On the other hand, Portugal was involved inthe political swing to the right, which was also taking place in the USA, France, Italy and other countries.

With the coalition between the centre-right PSD and the right-wing Partido Popular, political prioritiesaim at reducing the deficit of the state budget. Policies against poverty, and the GMI in particular, wereagain the subject of public debate. From a formal point of view, however, their design remained much thesame, with some small alterations.

The main initial priority of the government’s political orientation was labour law, aimed at greaterflexibility in labour relations, which have been one of the main sources of social conflict to date. The secondmajor priority was the revision of the Organic Law on Social Security, which nevertheless reproduced the mainoptions of the previous law and only gave more room to the optional social insurance systems by reducingthe maximum ceiling for compulsory contributions to the public pension systems. The GMI was the subjectof heated ideological debate. In the first stage, the government planned to introduce deep changes, includingmore restrictive eligibility criteria. The final result was but a change of name, and doing away withautomatic annual renewal, replacing it with a renewal of the application subject to an annual reappraisal ofthe case.

The first, two-year NAPincl took place in the middle of a period of political transition. All the existingmeasures went ahead normally. The first NAPincl, as said before, also provided for the implementation oftwo other new priority policies. The first consisted of a political commitment to the concentration ofresources to eradicate poverty in the most problematic urban and rural areas involving multidimensional,integrated, coordinated action on the part of the different sectors, from housing to safety, from training toemployment, from education to cultural initiatives, and from infrastructure to the organization of thecommunity and social facilities.

The second was the extension of the concept of GMI insertion contracts to people who were notbeneficiaries of the measure, but who were still in need, i.e. all the beneficiaries of the social action services,with the twofold objective of providing them with the resources to solve their problems and transforming,modernizing and activating local social services, on the basis of a model of teamwork and work byobjectives that had been successfully implemented in the public employment departments under theNational Action Plan for Employment. It took advantage of the reorganization of the social security systemintroduced with the new basic law of 2000 to make this change, increasing the number of human resourcesavailable in the services and bringing them closer to the population. Neither of them was discarded in thesecond version of the NAPincl introduced by the new government in July 2003, for the period 2003–05. Onthe contrary, the 2003–05 plan does the following:

• maintains the goals of reducing poverty, although they are a little more limited (or cautious) than in theprevious one;

• adds to the main challenges of the first version the objectives of activating the long-term unemployed,establishing a target of ongoing training and qualification of young people with little schooling;

• also includes as one of its big major challenges increasing minimum pensions to bring them up to thelevel of minimum wage by 2006;

• commits to a series of new goals in the areas of childhood, protection of children at risk, and the fightagainst school dropouts;

• maintains as a priority the “new” measure of the NAPincl for contractualizing insertion plans for allcases of exclusion.

LUÍS CAPUCHA ET AL. 199

In terms of smaller challenges, such as operational programmes, it reinforces the role of the social network,replacing the minimum income CLA, which involve the same partners, in order to simplify procedures. Ithas created the PROGRIDE programme to replace the National Programme against Poverty,64 which, withthe resources from the structural funds, will make it possible to develop urban social development contractsand the Rural Areas and Social Development programme.

In view of this continuity and development, one may well ask whether the political orientation of the new,right-wing, neo-liberal government has had any significant influence on the growth of social policies, thefunds allocated to them, or the modernization of the political model for activating subjects, as has beenhappening all over the rest of Europe. The answer is ambivalent, currently reflecting a stalemate.

On one hand, and from a formal standpoint, the rules and institutions of the new generation of activesocial policies designed to fight against poverty remain practically intact. From a practical point of view, onthe other hand, the level of performance has suffered in almost all fields not cofinanced by the structuralfunds. Thus, while the family-oriented discourse emerges again, investment in social family-supportfacilities has been practically reduced to zero. The supply of additional human resources needed toimplement the active policies has been cut off or reduced and the modernization of the administration ofsocial action has been suspended. Where an improvement in quality was expected in the resources availablefor local intervention, the budget has been reduced. The budget of the new PROGRIDE programme is lowerthan the total budget of the previous programme. The initiatives in the field of labour law created a newclimate of social discontent, which interrupted the implementation of the agreements on education, trainingand employment and the change or participation of the social partners in the management and assessment ofpolicies.

There is currently an impasse between the legal framework that constitutes a social state using activemeasures in a context of investment and innovation in inclusion policies, which are those that are still in theNAPincl, and a state of more liberal tendencies with no power to change the legal framework and replace itby the concept of privatization of services, but still capable of “devitalizing” it in the name of the tripleobjective of more flexible labour relations, family re-privatization of social services and, above all, areduction in expenditure—including social investment—in order to reduce the budgetary deficit of thestate.

Table 5.11 Chonology of legislation on poverty alleviation in Portugal

Year Legislative Act Content

1934 Law no. 1884, 16 March Institutionalizes the Social Security System as a compulsory systemof social insurance, composed of independent institutions by thestate. The system covers the risks of illness, disability, old age anddeath.

1962 Law no. 2116, 18 June Implements a welfare reform establishing State responsibilities inregulation and strategic co-ordination between Social Security,Health and Assistance.

1669 Law no. 2144, 29 May Creates the Regime of Social Protection for Agricultural Workers,covering maternity and family allowances, illness, disability, oldage and death.

1973 Decree-Law no. 142, 31 March Approves the legal framework of the Survival Pension.

1974 Decree-Law no. 217, 27 May Creates the Social Pension.

1975 Decree-Law no. 169-D, 31 March. Approves the legal framework of the Unemployment Benefit.

1976 Constitution of the Portuguese Republic Creates the Social Pension.

200 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Year Legislative Act Content

1977 Decree-Law no. 724, 12 DecemberPortaria no. 162, 24 March.Portaria no. 115, 9 March

Creates the 13th month payment for pensioners.Approves the legal framework of the Occupational Hazard benefit.Creates the Regime of Social Protection for the Self-Employed.

1978 Decree-Law no. 180-C, 15 July Creates the Regime of Social Protection for Domestic Work.

1980 Decree-Law no. 464, 13 October Defines new eligibility criteria for the Social Pension.

1982 Decree-Law no. 8, 18 JanuaryDecree-Law no. 284, 22 July

Amends the legal framework for the Self-Employed.Amends the legal framework of Domestic Workers.

1983 Decree-Law no. 297, 24June Defines new eligibility criteria for the Unemployment Benefit.Amends the legal framework of the Regime of Social Protection ofAgricultural Workers.

1984 Law no. 28, 14 AugustLaw no. 4, 5 April

Establishes the Social Security Framework Law, which sets out theoverall system at three levels: General Regime, Non-ContributiveRegime and Social Assistance. Creates Social Protection inMaternity, Paternity and Adoption.

Year Legislative Act Content

1989 Law no. 9, 2 May Establishes the Framework Law ofPrevention, Rehabilitation andIntegration of People withDisabilities.

1990 Resolution of Council of Ministersno. 8, 20 March

Creates the National Programme toFight Poverty (PNLCP) in the socialassistance subsystem.

Portaria no. 470, 23 June Creates the 14th month payment forpensioners.

1992 Resolution of Council of Ministersno. 189, 18 August

Creates the Programme for Supportof Families and Children (PAFAC) inthe social assistance subsystem.

1993 Decree-Law no. 328, 25 September Amends the legal framework of theRegime of Social Protection of theSelf-Employed.

1994 Law no. 34-B, 27 DecemberDespacho Conjunto MS/MESS, 20July

Creates the IVA Social, allowing for apercentage of the value-added tax tobe transferred to the Social SecurityBudget.

Creates the Programme of IntegratedCare for the Elderly (PAII) in thesocial assistance system.

1996 Law no. 19-A, 29 June Creates the Guaranteed MinimumIncome, implementing a benefit ofthe noncon tributive regime of socialsecurity, and a programme of socialinsertion.

1997 Resolution of Council Ministers no.197, 18 November

Creates the Social Network, astructural programme and essentialtool in the process of localdevelopment. Establishes the

LUÍS CAPUCHA ET AL. 201

Year Legislative Act Content

implementation of municipally-basedstrategic planning processes as afoundation for social intervention.

1999 Law no. 100, 13 September Amends the legal framework of theOccupational Hazard benefit.

Law no. 147, 1 September Establishes the Framework Law forthe Protection of Children and Youthat risk.

Decree-Law no. 119, 14 April Amends the legal framework ofUnemployment Benefit.

Decree-Law no. 9, 8 January Introduces flexibility in retirementage.

2000 Law no. 17, 8 August Establishes the Solidarity and SocialSecurity Framework Law.

2001 Resolution of Council of Ministersno. 91, 6 August

Approves the National Action Planfor Social Inclusion,

2002 Decree-Law no. 35, 19 February Defines new rules for the PensionBenefit Formula, in the context of theSolidarity and Social SecurityFramework Law.

Appendix 2

Table 5.12 Main benefits in the social protection system in Portugal (2000)

Risk Benefit General regime Non-contributory

employees self-employed

Family allowances Family allowance for children andyoung people

b o b

Family allowance supplements forhandicapped children and young people

b o b

Special education allowance b o b

Monthly living allowance b o –

Allowance for the assistance of a thirdparty

b o b

Funeral allowance b o –

Sickness Sickness benefit b o –

Christmas and holiday bonus bo

Maternity benefits Benefit in case of particular risksduring pregnancy

b b –

Pregnancy allowance b b –

Maternity, paternity and adoption leaveallowance

b b

202 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Risk Benefit General regime Non-contributory

employees self-employed

Special leave to assist children who areseriously handicapped or sufferingfrom a chronic illness

b – –

Assistance for sick or handicappedminor children

b – –

Parental leave allowance b b –

Special leave for grandparents b b –

Father leave of 5 days after the childbirth

b b –

Occupational illness Increased pensions for permanentincapacity

b o –

Temporary incapacity benefit b o –

Permanent incapacity pension b o –

Capital sum on death b o –

Housing rehabilitation allowance iftotal permanent incapacity

b o –

Death grant and reimbursement offuneral expenses

b o –

Risk Benefit General regime Non-contributory

employees self-employed

Occupational illness contd. Cash redemption for permanentincapacity

b o –

Provisional pension b o –

Supplementary care benefit b o –

Christmas bonus and holiday bonus b o –

Vocational training courses attendanceallowance

b o –

Unemployment Unemployment benefit b – –

Social unemployment benefit b – –

Partial unemployment benefit b – –

Invalidity Invalidity person b b b

Long-term care supplement b b b

Pension supplement for dependantspouse

b b –

Old age Old-age pension b b b

Long-term care supplement b b b

Pension supplement for dependantspouse

b b

Death Survivor’s pension b b b

Death grant b b –

LUÍS CAPUCHA ET AL. 203

Risk Benefit General regime Non-contributory

employees self-employed

Long-term care supplement b b b

Guaranteeing sufficient resources Minimum Income Scheme – – b

Source: MTS (2003) adapted.Notesb benefit,o optional benefit.– no benefit.

Table 5.13 Main schemes in the social protection system in Portugal (2000)

Social action Measures undertaken

Network of facilities andservices for children Nursery and kindergartenFamily nurseriesDay care centresHomes for children and young peopleShelter centresCommunity centres

Drug addiction Street teamsDay care centresResidential communities and reinsertion apartmentsCanteens

People with disabilities Homes (residência) for persons over the age of 16 years withpermanent or temporary disabilities.Homes for supported life (unidade de vida apoiada) for persons withpermanent mental disabilities.Homes for protected life (unidade de vida protegida) for adults withserious, probably becoming permanent, psychological problems.Homes for autonomous life (unidade de vida autónoma) for adultswith serious, probably becoming permanent, psychological problems,but who maintain a certain degree of autonomy.Sheltered workshops (centro de actividades ocupacionais) forseriously handicapped persons.Centres for social and occupational measures (forum socio-ocupacional) for persons with a light mental handicap.Nursing homes for temporary stay (lar temporário) of disabledchildren and young persons between the ages of 6 and 16 years.Premature intervention s(intervenção precoce) integrated aid measurecombining education, health and social action for children up to 6years old with disabilities or with a serious risk of backwardness.

Network of facilities and services for old age Home care (ajuda domidliária)Foster families (famílias de acolhimento integrado)Nursing homes for the permanent stay (lar permanente) of elderlypersons who are or risk to become severely dependent.Temporary Reception Centre for Emergencies(Centro de Acolhimento Temporário de Emergência) for old peoplein difficult social situation.Night centre (centro de noite) for old isolated people in need of aidduring the night.

204 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Social action Measures undertaken

Day care centres (centro de dia) for the elderly.

Notes

1 The research project on which this chapter is based was coordinated by Luís Capucha.2 Most common expression to describe the period of dictatorship.3 According to the first study on poverty in Portugal, published in 1984 and based on data from the first National

Survey of Family Expenditure in 1973/74, using a poverty line of 75 per cent of the average of the adultequivalent of family expenses per year (the threshold of 50 per cent situated the relative poverty line under theabsolute poverty line, the reason why a threshold of 75 per cent was used), 43.2 per cent of individuals and 43.4per cent of families were poor at the beginning of 1974. The rate was higher in urban areas (45.7 per cent) than inrural areas (42.5 per cent) (Silva 1984). Agricultural workers, rural property-owners, non-agricultural manualworkers and tradesmen were the most affected socio-occupational categories. Among the non-workingpopulation the poverty rate rose to 59 per cent. On the basis of the same data it has been estimated that about 28.2per cent of urban families and 38.9 per cent of rural families lived in absolute poverty (Costa et al. 1985).

4 Including tax and social security contribution evasion, as well as family-based agriculture, personal careprovision and many other activities in other sectors.

5 This figure corresponds to a poverty threshold of 75 per cent of the average value for family income per adultequivalent in 1980, calculated on the basis of the Survey of Household Income and Expenditure for 1980/81. Thedistribution of poverty by socio-occupational groups remained identical to that recorded in 1974. Another studydrawing on the same statistical source but using poverty lines of 50 per cent, 66 per cent and 75 per cent of themedian income per adult equivalent showed that the poverty rates were 11.5 per cent, 24.2 per cent and 31.2 percent respectively for those thresholds (Ferreira 1993). Using a methodology based on a diet, the same authorcalculates that absolute poverty was around 24.1 per cent of individuals in 1980 (Ferreira 2000). The differencein the results of the two studies is explained by the different methodologies used.

6 The First Republic is the name given to the democratic regime that followed the republican revolution in 1910. Itlasted until 1926, when a coup d'état replaced it with the Estado Novo, of a fascist, authoritarian nature.

7 Institute for Compulsory Social Insurance.8 Decree Law no. 23048, 23 September 1933.9 National Labour Statute, the basic legal framework of the Portuguese corporative regime.

10 On a very limited basis, given the scarce availability of the under-developed health system services.11 Corporative House of Parliament.12 Fishermen’s funds.13 According to the same data, social security receipts represented 3.94 per cent of GDP, being 2.74 per cent from

contributions, 0.48 per cent from the State Budget and 0.72 per cent from other transfers (Carreira 1996:96).14 Law no. 2115, 18 June 1962.15 National corporative unions’ welfare funds, which were responsible for social protection and family allowance

schemes.16 Federation of Welfare and Family Allowance Funds.17 National Pensions Fund.18 Ministerial Order no. 21546, 23 September 1965.19 Insurance Funds.20 Including a decrease in contribution periods, an increase in the birth, death and marriage benefit, a new benefit

formula for pensions, and the inclusion of domestic work (Serrão 2000).21 When their status as head of the family was not recognized.22 Receipts rose from 3.94 per cent of GDP to 9.4 per cent in the same period. In 1975 contributions represented 8.

11 per cent of GDP and transfers from state budget 0.80 per cent.

LUÍS CAPUCHA ET AL. 205

23 Law no. 56, 15 September 1979.24 Social Security Financial Management Institute.25 Social Security District Centres.26 Some substantial alterations in this field came later, at the end of the 1980s when the fourteenth-month benefit

was introduced for pensions.27 Eurostat, Newcronos, 1 April 2004.28 Despite the existence of scattered social assistance—to face situations where there was a risk of social breakdown

—policies to combat poverty were extremely limited in scope and often applied arbitrarily. On the other hand,with poverty as with other risks, the family was the centre for preventing social emergencies, but it was also onefor creating vicious closed circles of intergenerational poverty, e.g. the intergenerational transmission of povertysituations.

29 Council of Ministers Resolution no. 8, 20 March 1990.30 Another study, based on the same data, measured poverty using expenses, the OECD equivalence scale and a

poverty line evaluated in terms of basic needs, and displayed a poverty rate of 22.3 per cent in 1989/90 (Costa1993).

31 We preferred to use Leonor Vasconcelos Ferreira’s study in previous paragraphs because it puts in evidence thecomposition of poverty. Unfortunately, the author did not explore the 1994/95 Household Budget Survey. Theperiod 1990–95 can be covered, as far as the main indicators of poverty and income inequality are concerned, byusing another approach to the same data, giving rise to different results, but not affecting the main point of thewidespread nature of poverty in Portugal nor the maintenance of regular poverty levels between 1980 and 1995,in spite of the registered oscillations.

32 Difference between the median equivalized income of persons below the poverty threshold and the thresholditself, expressed as a percentage of the threshold itself.

33 In 2000, there were 629,000 pensioners under the general scheme with a minimum pension. The value of thispension was little more than half the minimum salary, which in turn was only slightiy higher than the averagepension. However, the minimum pension has risen more than 23 per cent since 1995.

34 Those � 75 years old.35 A systematic comparison between the Portuguese Household Budget Survey 1994/95 (PHBS) and European

Household Panel 1995 (ECHP) can be found in Rodrigues (1999). The difference of poverty rates, using similarformula, is about 3 percentage points higher in the ECHP if only monetary income is considered in PHBS, andalmost 6 points when total income is taken into consideration. Also severity and intensity are considerably lowerin PHBS.

36 Measured in terms of relative poverty, defined as the percentage of the population living on an income of lessthan 60 per cent of the national median of equivalent income after social transfers.

37 Eurostat, Newcronos, 3 December 2003.38 INE, Inquérito ao Emprego.39 Eurostat, Labour Force Survey.40 Eurostat, Newcronos, 24 July 2003.41 INE, Contas nacionais. Forecasts up to 2002 pointed to a reduction in this figure to about 62.7 per cent. 42 Although there are no official estimates in this regard and no studies have been published on the subject since the

early 1980s, the value of the informal economy is believed to vary between 23 per cent and 25 per cent of theGDP. It includes activities from agriculture and self-supply to crime, tax fraud, evasion of social securitycontributions, illegal employment and similar situations.

43 Several studies show that it is among the most qualified people and high incomes that a slight improvement in thedistribution of domestic responsibilities between men and women is taking place (Torres et al. 2000).

44 Law no. 17, 8 March 2000. See Appendix 2 for a more detailed description of the system.45 In the previous scheme the calculation was made upon the best ten years from the past fifteen of contributions,

encouraging evasion.46 The 2001–03 NAPincl defined inclusion as

206 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

a dual process of transforming structures and social, political, economic, and cultural institutions so asto include people on the basis of their specific needs and guarantee their rights, creating the necessaryopportunities and empowerment conditions to enable them to fully assume their duties and responsibilitiesto themselves, their families and the community to which they belong.

47 A CLA consists of the municipal council (and, in principle, is presided over by the Mayor), parish councils,central public administration organizations established in the area and private non-profit entities. The CommunitySupport Fund may comprise the parish councils (and, in principle, are presided over by their chairpersons),central public administration organizations established in the area, other private non-profit entities andrepresentatives from social groups of importance in local intervention.

48 A policy consisting of integrated training addressing the plurality of personal, social and occupational skills,according to individual plans that range from occupational orientation to practical training and support toinsertion in the labour market. This kind of intervention was first developed in the field of rehabilitation ofdisabled people and then adapted to other groups and categories of trainees with the support of Horizon 1Programme.

49 The measure is supported by the Community Support Fund (CSF) II and III, and the European Social Fund.50 For instance in 2000 there were 44,631 long-term unemployed in occupational programmes, 5,288 youngsters in

craft school workshops, 3,109 people in social companies and 677 disabled persons in protected employment,according to the National Action Plan for Employment. The actions undertook under the special traininginitiatives have involved 12,000 youngsters in risk, homeless people, ex-drug addicts, ex-prisoners, and personsbelonging to ethnic minorities in 2002, and also 7,000 disabled persons, numbers estimated from the evaluation ofOperational Programme Employment, Training and Social Development (CSFIII).

51 The stability agreement encompassed monetary and fiscal policy constrains and currency devaluations. Policiesaimed in a first stage at retrenching domestic demand with the aim of reducing imports and increase exports. Thisinvolved a raise in interest rates, limits to the expansion of credit in the economy, setting of salary cut-offs,increase of the tax burden, and a restrain of the financing to the public sector. In the second phase, currencydevaluations took place in order to allow Portuguese goods and services to become more competitive at theinternational level (Lopes, J.da Silva 1996:146–148).

52 This amount corresponds to 36 per cent of the minimum wage. 53 This is intended to stimulate activation and prevent dependence.54 GMI Local Monitoring Committee, responsible for the local (municipal) management of the scheme.55 All members over the age of 16 must sign. If a member does not comply, s/he is withdrawn from the household

for the purpose of calculating the benefit, though his/her income is still considered, thus adversely affecting thetotal benefit. If the beneficiary fails to comply, the benefit is cancelled and a penalty is imposed under which s/heis unable to submit another application for the next six months.

56 IDS was a department of the Ministry of Labour and Solidarity (it was merged into the Institute of Solidarity andSocial Security in 2003). It was responsible for developing and co-ordinating social policies, for supporting thedevelopment of partnerships between key stakeholders, and for the co-ordination of the NAPincl and theparticipation in the European Inclusion Process.

57 Which represents an additional innovation: a national measure being managed at local level.58 In 1995 56 per cent of the poor in Portugal were concentrated in these regions (Carvalho 2000).59 “Other reasons” include alterations in the household members, death, imprisonment for more than two years,

moving away from the district, a request from the applicant and other unspecified reasons.60 The evaluation of the experimental phase of the measure acknowledged that, despite the insertion programmes,

the simple fact that there were several families starting to receive a stable, regular income would lead to theintroduction of a factor of strategic rationality in attitudes and independent efforts towards activation, as well asan improvement of their living conditions.

61 This is the age of entitlement for old age and social pension.

LUÍS CAPUCHA ET AL. 207

62 It is worth noting that the amount of the cash transfer, being so small, has not been a topic of heated politicaldebate.

63 Family allowance levels vary inversely to household income levels according to three predefined income cohorts.64 See p. 230 for a full description of the Programme.

References

AA. VV.(1999) A Acção Social em debate, Lisbon: DGAS.Almeida, A.N. et al (2000) “Family relations: change and diversity”, in J.Viegas and A.F.Costa (eds), Crossroads to

Modernity: Contemporary Portuguese Society, Oeiras: Celta, pp. 41–70.Almeida, J.F. et al. (1994) Exclusão Social: Factores e Tipos de Pobreza em Portugal, Oeiras: Celta.Beck, U. (1992) Risk Society, London: Sage.Capucha, L. (2000) “Poverty and social exclusion”, in J.Viegas and A.F.Costa (eds), Crossroads to Modernity:

Contemporary Portuguese Society, Oeiras: Celta, pp.179–205.Capucha, L. et al. (1998) Rendimento Mínimo Garantido: avaliação da fase experimental, Lisbon: CIES-MTS.Carreira, H.M. (1996) As políticas sociais em Portugal, Lisbon: Gradiva.Carvalho, M.F. (2000) Distribuição do rendimento e pobreza em Portugal e suas regiões, Documento de Trabalho,

Lisbon: DPP.CES (1997) Pareceres sobre o Rendimento Mínimo Garantido, Lisbon: CES.Clijsters, E. (1992) “Portugal 1974: a non-violent revolution: causes, course and consequences of the ‘Revolution of

Carnations’, confronted with theoretical definitions of revolution”, PhD thesis, European University Institute,Florence.

Comissâo Nacional do Rendimento Mínimo (1998a) Comunidades ciganas e inserção social no âmbito do RMG,Encontro Reflectir para Agir, Lisbon: Policopiado.

Comissâo Nacional do Rendimento Mínimo (1998b) Manual para a inserção, Lisbon: IDS.Comissão Nacional do Rendimento Mínimo (1999) Relatório de Avaliação da Actividade das CLA, Lisbon: IDS.Commission of the European Communities Directorate General Employment, Industrial Relations and Social Affairs

(1991) Research Problematics: Feasibility Studies, Geie Animation and Research, Lille.Costa, A., Bruto da Silva, M., Pereirinha, J. and Matos, M. (1985) A Pobreza em Portugal, Lisbon: Caritas.Costa, A.Bruto da (1993) “The paradox of poverty, Portugal 1980–89”, PhD thesis, University of Bath.DEPP/MTS (2000) Carta Social, Lisbon: Celta Editora.DEPP/MTS (2002) Portugal 1995–2000 Perspectivas da Evolução Social, Lisbon: Celta Editora.EC (1999) MISSOC—Social Protection in the Member States of the European Union: Situation on 1 July 1998 and

Evolution, Luxembourg: Office for Official Publication of the European Communities.European Communities (2001) European Social Statistics: Social Protection Expenditure and Receipts, 1980–99,

Luxembourg: Office for Official Publications of the Euro pean Communities.Eurostat (1990) Poverty in Figures, Report prepared by R.Teekens and A.Zaidi of Institute of Social Studies Advisory

Service (ISSAS) for Eurostat, Luxembourg: Office for Official Publications of the European Communities.Ferreira, L.V. (1993) A Pobreza em Portugal—variação e decomposição de medidas de pobreza a partir de arçamentos

familiares de 1980/81 e 1989/90, Lisbon: CISEP.Ferreira, L.V. (2000) A pobreza em Portugal na década de 80, Lisbon: Conselho Económico e Social.Ferrera, M., Matsaganis, M. and Sacchi, S. (2002) “Open coordination against poverty: the new EU ‘inclusion

process’”, Journal of European Social Policy, 12 (3):227–239.Fields, R.M. (1975) The Portuguese Revolution and the Armed Forces Movement, New York: Praeger.Giddens, A. (1991) Modernity and Self-identity: Self and Society in the Late Modern Age, Cambridge: Polity Press.Gouveia, M. and Rodrigues, C.F. (1999) The Impact of a Guaranteed Minimum Income Programme in Portugal, WP 3/

1999, Lisbon: DE/ISEG.

208 PORTUGAL: PATH TOWARDS MINIMUM INCOME?

Graham, L.S. and Wheeler, D.L. (eds) (1983) In Search of Modern Portugal: The Revolution and its Consequence,Madison, WI: University of Wisconsin Press.

Guerreiro, M.D. and Lourenço, V. (1999) Emprego, Familia e Actividades Comunitárias: uma relação maisequilibrada para homens e mulheres—0202 Project—Euro pean Foundation for the Improvement of Living andWorking Conditions— Report on Portugal, Lisbon: CIES/ISCTE.

Guibentif, P. and Bouget, D. (1997) As políticas do Rendimento Mínimo na União Europeia, Lisbon: União dasMutualidades Portuguesas.

Guillén, A. and Silva, P.A. (2001) Redesigning the Spanish and Portuguese Welfare States: The Impact of Accessioninto the European Union, Working Paper no. 85, Florence: Centre for European Studies.

Harsgor, M. (1976) Portugal in Revolution, Beverly Hills, CA: Sage Publications.Harvey, R. (1978) Portugal, Birth of a Democracy, London: Macmillan.IDS (2000) Execução da Medida e Caracterização dos beneficiários RMG, Lisbon: IDS.IDS/MTS (2000) Políticas Sociais Activas em Portugal, Lisbon: IDS.Kayman, M.A. (1987) Revolution and Counter-Revolution in Portugal, London: Merlin Press.Larsson, A. (2002) “The new Open Method of Co-ordination—a sustainable way between a fragmented Europe and a

European supra State?—a practitioner’s view”, Lecture at Uppsala University.Lopes, J.da Silva (1996) A Economia Portuguesa desde 1960, Lisbon: Gradiva.Magone, J.M. (1997) European Portugal: The Difficult Road to Sustainable Democracy, New York: St Martin’s Press.Maia, F. (1985) Segurança Social em Portugal: Evolução e Tendências, Lisbon: Instituto de Estudos para o

Desenvolvimento, 1st edn.Mozzicafreddo, J. (1985) “Etat, mouvements et luttes sociales—processus politique portugeais 1974–76”, PhD thesis,

University of Montpellier 1, France.Mozzicafreddo, J. (1997) Estado-Providência e Cidadania em Portugal, Oeiras: Celta Editora.MSA (1965) Seminário Desenvolvimento Comunitário, Lisbon: Instituto de Assistência a Família.MTS (2001) Plano Nacional de Acção para a Inclusão Portugal 2001–2003, Lisbon: MTS.Pedroso, P. (1997) Rendimento Mínimo Garantido: Ideias, experiências e desafios para as políticas sociais em

Portugal, Lisbon: Conselho Económico e Social.Pereirinha, J. (ed.) (1991) Observatory on National Policies to Combat Social Exclusion, Lille: DGV, EEIG.Rodrigues, C.F. (1993) Measurement and Decomposition of Inequality in Portugal (1980/81–1989/90), Lisbon: CISEP.Rodrigues, C.F. (1999) Income Distribution and Poverty in Portugal (1994/95): A Comparison between the European

Community Household Panel and the Household Budget Survey, Lisbon: CISEP, ISEG/Universidade Técnica deLisboa.

Rodrigues, C.F. (2001) Anti-poverty Effectiveness and Efficiency of the Guaranteed Minimum Income Programme inPortugal, WP 8/2001, Lisbon: DE/ISEG.

Rodrigues, M.J. (ed.) (2000) Para uma Europa da Inovação e do Conhecimento: emprego, reformas económicas ecoesão social, Oeiras: Celta Editora.

Room, G. et al. (1993) Observatoire européen sur les politiques nationales de lutte contre l’exclusion sociale, Lille:DGV, EEIG.

Santos, B.S. (1990) O Estado e a Sociedade em Portugal (1974–1988), Porto: Afrontamento.Schmitter, P.C. (1975) Corporatism and Public Policy in Authoritarian Portugal, London: Sage Publications.Serrão, J. (2000) Dicionário de história de Portugal, Porto: Livraria Figueirinhas.Silva, M. (1984) “Uma estimativa da pobreza em Portugal, em Abril de 1974”, Revista de Ciências Sociais, 1, June:

117–128.Torres, A.C., Monteiro, L.T., da Silva, V. and Cabrita, M. (2000) Men and Women between Family and Work in

Portugal, WP 09/2000, Lisbon: WORC. Available online www.seg-social.pt.

LUÍS CAPUCHA ET AL. 209

6Poverty and the safety net in Eastern and South-Eastern

Europe in the post-communist eraDimitri A.Sotiropoulos

For the next few years, Central and Eastern Europe will be facing a crisis in the development ofsocial protection, in the classic Greek sense of both a threat and an opportunity. The threat isthat there will be persistently high levels of poverty, accompanied by wide inequality andsocioeconomic fragmentation; the opportunity is still there to create a new system that builds onthe sense of social solidarity and desire for redistributive justice that are still prevalent in theregion.

(Standing 1996:251)

Introduction

In studies of welfare reform in post-communist Eastern Europe, it is customary to argue that, unless there isan improvement in the structure and administrative capacity of the involved states, little progress towardsreform is expected. This is so because after all a “welfare state” is mainly a state, that is, a set ofadministrative institutions. These institutions are in charge of poverty relief, uplift out of social exclusion,insurance against risks and redistribution over the life cycle (Barr 2001a:1). In other words, in addition tobeing a regime of variable de-commodification and a system of social stratification (Esping-Andersen 1990:20–21, 23), the welfare state is also a set of organizations (Skocpol 1992:42–43). In order to analyse theparticular kind of social protection one is interested in, it is necessary first to answer the question what typeof state organization one speaks about.

With respect to welfare provision, the pre-transition states of Eastern Europe have been aptly called“caring patronage states” (Bauman 1993: 20, 22) or “bureaucratic collectivist” systems (Deacon 1992). Theformerly socialist welfare systems, despite their comprehensive character, had nothing to do with thecomprehensiveness of social protection found in the North European social democratic model. They wererather statist and hierarchical (Ferge 1992:207), privileging members of the nomenclature and workers overother social strata. At the price of restricted political freedoms, democratic un-accountability and extendedideological control, the residents of former socialist countries lived under the caring tutelage of strong andomnipresent states (Sotiropoulos 2002). Such states offered wide, albeit uneven and generally low-quality,social protection.

The nature of post-communist welfare states is less clear. Such systems are still in the making and theydo not belong to any one type or “welfare regime” (Ferge 2001:12–13). In the beginning of the transition,they only reluctantly shed their centrally-planned statist features and some of them even expanded in size(first phase of transition from socialism). Towards the end of the decade of the 1990s several of the post-communist welfare states started making a systemic shift toward a more differentiated, three-pillar pension

system (second phase of transition). This trend can be explained by the pressure of internal, fiscalconstraints, felt by post-communist governments, as well as by the influence of international agencies inmatters of welfare state reform (Standing 1996:230; Deacon 1997; Ferge 2001). Yet the legacy of socialistwelfare states has shaped and will shape post-transition social policy-making long after the collapse of EastEuropean socialism (Ringold 1999:15).

The outline and data sources of this chapter

In what follows, we will first summarize the historical legacies of social protection during socialist regimesand during the first phase of transition from socialism. We will then briefly survey sub-regional differencesin the extent of poverty and income inequality in Eastern and South-Eastern Europe. In this chapter, under“Eastern Europe”, we include all former socialist countries located between Central Europe and theeasternmost areas of the former Soviet Union. Under “South-Eastern Europe”, we include all Balkancountries, except for Greece and Turkey. We are going to present a short overview of the causes of incomeinequalities and poverty and list poverty alleviation measures. We will illustrate this presentation with thecases of Bulgaria and Romania, discussed in a separate section. We will then discuss available strategies forthe future, referring to the possible impact of the EU, to comparable lessons to be drawn from theexperience of Southern Europe, i.e., of Greece, Italy, Portugal and Spain and to the views of the WorldBank. We will end with a short note on who may be the welfare state reformers and the suitable socialpolicy mix.

Data used in this chapter are sensitive to official poverty measures and national definitions of poverty,adopted by each state. For instance, only 2 per cent of all Croatian households fall below the officialCroatian poverty line (World Bank 2000:295). This is hard to believe given that fourteen years ago, at thevery time point of transition, Croatia used to have the most unequal income distribution among all East andSouth-East European countries (World Bank 2000:140, Table 4.1). On the other hand, cross-sectionalcomparisons of poverty, based on a fixed dollar price, are also disputable when it comes to economies withvery variable consumer prices indices. For these reasons, we will resort to both international and officialnational data and treat our conclusions with caution.

Historical legacies of the socialist era and the first phase of transition

The socialist welfare state

In terms of income inequality, before the transition the societies of Central and Eastern Europe, including thevast USSR, were among the most equal in the world. This was due to the comparatively small differences insalaries and wages, to the wide, albeit uneven, social protection covering the population as a whole, and tothe rather elevated status of manual workers.

There was a basic sense of security felt by almost all members of socialist society. All workers, men andwomen were entitled to full-time employment, to the point that firms used to keep redundant workers on theirpayrolls. Officially, there was no unemployment, since the state guaranteed a job to all citizens. Workerswere also entitled to social security benefits, dispensed at the firm-level. The link between employment andsocial benefits was provided by the firm which acted as the functional equivalent of a local social securityadministration (Wagener 2002:155). Firms functioned on “soft budget” constraints, meaning that the statewould compensate for any deficits incurred by firms. Workers did not pay contributions out of their wagesto pension schemes, but firms did so out of their budget.

DIMITRI A.SOTIROPOULOS 211

There was a very high participation rate of women in the labour force, who were thus entitled to the sameworking and social security compensation as men. In addition, if unable to work, pregnant women andmothers with young children were paid maternity and family benefits (Lipsmeyer 2000:1206). Educationwas provided for free and funded by the state on the basis of general taxation. Health care was free too, eventhough entitled citizens resorted to bribes or connections with the individual members of the authorities, inorder to obtain either priority in or higher quality of medical treatment. Health care was wholly funded bythe state, through tax revenues. In addition, the prices of goods, such as foodstuffs and clothes, and services,such as communications and transport, were heavily subsidized by the state, offering an additional relief tofamily incomes.

Poverty as a social problem was either unknown or well dressed-up in Eastern and South Eastern Europebefore the disintegration of socialist regimes. In view of the above, a West European type of safety net, inthe form of targeted benefits to the victims of the capitalist market, was considered unnecessary undersocialism in Eastern Europe.

However, all was not quiet on the socialist welfare front. There were significant wage differentials andsegregation along gender lines (Standing 1996:227, 229: Emigh et al. 2001:9–10). Some categories of thepopulation, such as Polish private farmers or East German small entrepreneurs, were not treated on a parwith workers of the socialist states (Wagener 2002:155). The living standards in the south-east Republics ofthe former USSR left much to be desired. But such disparities were nothing compared to the size ofinequalities which would emerge as a result of the collapse of socialism.

Policy shifts in the wake of the transition from socialism

In the first phase of the transition from socialism, some countries became early reformers of the welfarestate (e.g., Poland, Hungary), while others were laggards (e.g. the South-East European countries).However, overall, sooner or later similar structural and policy shifts were made across the board. The focalpoint of policy shifts was the firm. As firms closed down or started laying off redundant workers,governments shaped programmes for the unemployed. Early on in the transition, unemployment benefitswere generous, with sufficient replacement rates and duration periods (Lipsmeyer 2000:1198). In the realmof pensions, governments introduced pay-as-you-earn (PAYE) systems, funded by employers andemployees, but financially supported by state budgets. In the wake of the transition, firms stopped providingclinics for their employees. Governments introduced contribution-based systems in health care provision.There was an exception to the general trend of welfare reform: most post-communist governments optednot to alter either the structure or the generosity of maternity and family benefits. The reason for thisexception was that decision-makers reckoned that changes in the relevant programmes, coupled with risingunemployment, would endanger the legitimacy of the new democratic regimes or at least the chances of theincumbent governments getting re-elected (Lipsmeyer 2000:1200, 1203, 1206).

Yet, after 1989, there was sharp increase in income inequalities and poverty (Standing 1996:231, 235;Elster et al. 1998:220, 222, 246; Atal 1999:5–6, 8; Cox and Mason 1999:126–127, 131–133; Ringold 1999:20; Deacon 2000:148–149; Hutton and Redmond 2000:8–12; Pascall and Manning 2000:246; World Bank2000:1, 141; Barr 2001a:242–243; Ferge 2001:12, 16–17). Why did this happen? All the aforementionedpolicy shifts and new measures did not mean that social security and the fight against poverty were amongthe top priorities of post-socialist governments. Quite to the contrary: as the role of the firm was changing inthe new economic system, governments took little care to shelter the population from the effects oftransition. Post-communist governments understood the problem, in a typical fashion, as a trade-off between

212 SOUTH-EASTERN EUROPE: POVERTY

efficiency and equity (Scholz and Tomann 1999:93, 96). The development of social assistance policy wouldincrease the latter, but dampen the former.

The effects were dramatic: as unemployment grew, there were fewer working people to contribute tosocial insurance schemes. Output also fell quite spectacularly during the first three years of the transition(Barr 2001a:243). Because of output decline, there was decreasing revenue from collected taxes. Thisdecline was not compatible with the increasing needs of the still state-funded social protection.

Finally, the introduction of a market system meant that many goods and services were no longersubsidized by the state. Consequently, basic commodity prices were freed and started rising. Inflation grewrapidly. To sum up, under those circumstances in the first phase of the transition, governments scrambled toput together unemployment benefit schemes and new institutions jumped to offer social protection. Therelevant efforts were haphazard, and a more systematic approach to constructing a new welfare model wasleft for later (Wagener 2002:157). As a consequence of all the above, poverty soared and income inequalitiesbecame very deep in almost no time. While this was a common pattern throughout the post-communistworld, there were big differences among regions and among states.

Inter-regional differences in income inequality and poverty

Variations of income inequality

Even before 1989, some socialist societies were more equal than other ones. The former socialist republicsof Caucasus and Central Asia were characterized by considerably higher income inequality than thesocialist countries of Central Eastern Europe (e.g., Czechoslovakia, Hungary. See Table 6.1). Livingstandards in the Caucasus and Central Asia, parts of which were known as the “Soviet Third World”, werelow. Living standards in Central Europe were much higher. The rest of socialist societies, the Balkans, theBaltics and the Western areas of the former Soviet Union (Western FSU), fell somewhere in between thesetwo ends of the range (with Croatia and Moldova being cases of higher inequality, as Table 6.1 indicates).

The same pattern was observed after the transition. However, always in terms of income inequality, therewere two important differences. First, as noted above, in all former socialist societies income inequalitybecame worse. And second, while income inequality was remarkable in Central Eastern Europe, it becamedramatic in the regions of South-Eastern Europe and the CIS. Inequality was very high in Bulgaria, theRussian Federation and Moldova (and even higher in Armenia and other countries of the easternmostFSU).

Table 6.1 Income inequality in Eastern and South-Eastern Europe, before and after the transition

Region and country Gini coefficients for income per capita

1987–90 1996–99

Central Eastern Europe

Czech Republic 0.19 0.25

Hungary 0.21 0.25

Slovenia 0.22 0.25

Poland 0.28 0.33

South-Eastern Europe

Albania 0.27

DIMITRI A.SOTIROPOULOS 213

Region and country Gini coefficients for income per capita

1987–90 1996–99

Bulgaria 0.23 0.41

Croatia 0.36 0.35

Macedonia, FYR 0.37

Romania 0.23 0.30

Baltics

Lithuania 0.23 0.34

Latvia 0.24 0.32

Estonia 0.24 0.37

CIS and Western FSU

Russian Federation 0.26 0.47

Ukraine 0.24 0.33

Moldova 0.27 0.42

Belarus 0.23 0.28

Caucasus and Central Asia

Armenia 0.27 0.59

Georgia 0.29 0.43

Kyrgyz Republic 0.31 0.47

Kazakhstan 0.30 0.35

Tajikistan 0.28 0.47

Turkmenistan 0.28 0.45

Sources: World Bank (2000: Table 4.1, p. 140).Notes“Western FSU” means the Western areas of the former Soviet Union. Income refers to disposable income—after-tax

earned income. Pre-transition data is based on Family Budget Surveys by Milanovic (1998), used in WorldBank (2000). Post-transition data is based on World Bank’s calculations on the basis of representativenational household surveys.

Variations of absolute poverty

Variations in income inequality were replicated by variations in poverty. We may employ a measure ofabsolute poverty used by the World Bank, to compare the extent of poverty across former socialist regionsand countries, approximately eight years into the transition period. This measure is the proportion of peopleliving on US$4.30 per person, per day. There are large differences between Central Eastern Europe on theone hand and the Russian Federation and South-Eastern Europe on the other (Table 6.2). While in theformer the incidence of poverty is small, in the latter it is very extended. In some cases, such as in Albaniaand Moldova, at least half of the population is absolutely poor.

By contrast, in the Czech Republic and in Slovenia, less than 1 per cent

214 SOUTH-EASTERN EUROPE: POVERTY

Table 6.2 Absolute poverty rates and GNP in dollars per capita (1998)

Region and country Percentage living on $4.30 1998 GNP in dollars

(Per person, per day) per capital (1996 PPP)

Central Eastern Europe

Czech Republic 0.8 12,197

Hungary 15.4 9,832

Slovenia 0.7 14,399

Poland 18.4 7,543

South Eastern Europe

Albania 58.6 2,864

Bulgaria 18.2 4,683

Croatia 4.0 6,698

Macedonia, FYR 43.9 4,224

Romania 44.5 5,571

Baltics

Lithuania 22.5 6,283

Latvia 34.8 5,777

Estonia 19.3 7,563

CIS and Western FSU

Russian Federation 50.3 6,186

Ukraine 29.4 3,130

Moldova 84.6 1,995

Belarus 10.4 6,138

Caucasus and Central Asia

Armenia 86.2 2,074

Georgia 54.2 3,429

Kyrgyz Republic 84.1 2,247

Kazakhstan 30.9 4,317

Tajikistan 95.8 1,040

Turkmenistan 34.4 2,875

Sources: World Bank (2000: Table 1.1, p. 35).Notes“Western FSU” means the Western areas of the former Soviet Union. The head count index of absolute poverty is

based on surveys which took place between 1995 and 1999. In Albania, the survey did not include the capitalcity, Tirana. Poverty rates are based on World Bank’s estimates, while GNP estimates are drawn from theBank’s “World Development Indicators”. Braithwaite et al. (2000:91) prefer to work with relative povertylines than with absolute ones. Their argument is that the former enhances comparability when comparingcountries with very different GDP levels. The argument is probably correct, but in this chapter we useabsolute poverty criteria, in combination with GNP per capita, for purposes of quicker understanding andbetter illustration of the problem under study.

of the population live under those conditions. In between these two extremes, we find “medium-poverty”cases, such as the Baltics and Ukraine. This pattern of absolute poverty coincides roughly with the variation

DIMITRI A.SOTIROPOULOS 215

of GNP per capita (in 1998) in the same regions. The poorest regions and countries mentioned above showthe lowest GNP figures. An exception is Russia which has a rather high GNP per capita for the rather largeshare of its people who live with US$4.30 or less per day (Table 6.2).

This “Russian exceptionalism” is accounted for by the fact that, outside the countries of the Caucasus andCentral Asia, the Russian Federation is by far the most unequal of all former socialist countries in terms ofincome inequality (Table 6.1). In sum, the extent of poverty differs from one country to another, eventhough the starting point (1989) and the conditions of departure (transition from socialism) may look thesame.

Causes of income inequalities and poverty in the post-communist era

A social class-based explanation of increased income inequalities

After so many years which have passed from the turning point of the transition, how can we account for thepersistent income inequalities in the post-communist world as a whole? On the one hand, a general, across-the-board, social class-based explanation looks plausible. Since the transition, social class divides havebecome much wider than in the pre-1989 period. This development is usually accounted for by the rise of anew dominant social class (Eyal et al. 1998), which, in terms of consumption patterns, has quickly elevateditself above from the rest of society. The core of the new class is composed of former managers of state-enterprises and businessmen who have made money in illegal ways (trafficking of drugs, weapons as wellas illegal, male and female workers). On the edges, the new class also includes a broad range of people(locally referred to as “the Mafia”, Standing 1996:251), ranging from traders in the informal economy tocriminals and local warlords. Such ways of generating income are obviously available to a tiny section ofthe population. The rest simply try to make ends meet.

On the other hand, explanations of patterns of inequality had better be country- or region-specific. Forinstance, outside the Caucasus and Central Asia (which are not the focus of our study), the sharpestincreases in income inequality have taken place in Bulgaria and the Russian Federation. In post-communistBulgaria and Russia, the new class has been termed “capitalists without capitalism” because there wasaccumulation of private wealth in the absence of a fully fledged market. By contrast, in Central EasternEurope we may speak of “capitalism without capitalists” (Eyal et al. 1998:5), in the sense that after thetransition there was no propertied bourgeoisie. The new upper class has emerged out of the formernomenclature. In some of the above cases, one may speak of a “captured state”, which has fallen in thehands of new ruling elites. In the beginning of the twenty-first century, it looks as if, in the minds of post-communist elites, the reduction of income inequalities is often one of the lowest priorities. This is a socialpolicy task which unfortunately could be taken up only to the extent that it does not impede either thepolitical elites’ electoral prospects or the macro-economic survival of post-communist economies.

Who are the poor in post-communist societies?

Which social categories and groups have been hit hardest by poverty and how does the typical poor personin post-communist societies look like? In Eastern Europe as a whole, the population categories whichinclude large numbers of poor people are the unemployed, children, the residents of rural areas and theminorities. The Roma (gypsies) are particularly hit by poverty (Emigh et al. 2001). All these are socialgroups with high rates of poverty, but are relatively small groups. They do not make up the majority of allpoor people.

216 SOUTH-EASTERN EUROPE: POVERTY

Barr distinguishes between two large categories of poor people in post-communist Eastern Europe. Thefirst category is composed of frail elderly and/or disabled people, while the second category is made up ofthe “new poor” (Barr 2001b: 181). The latter category is the largest. The majority among the poor is madeup of individuals who belong to the 15–64 age group, live in the cities and are employed (World Bank 2000:5). To put it in another way, in Eastern Europe increasingly the poor are the “working poor”. For instance,in the Russian Federation and Ukraine, there has been a growing number of low-paid unskilled workers whoare not entitled to benefits of the new social assistance schemes in place since the late 1980s (Standing 1996:232). In other words, the typical poor person is the “average citizen”.

How has the usually unexceptional, “average citizen” become a typically poor person? This is probablyrelated to the particular characteristics and causes of poverty in post-communist Europe, which set this caseof poverty apart from the incidence of poverty in the rest of the developed world.

In contrast with the pockets of poverty found in the West, in Eastern and South Eastern Europe povertyhas been very extended and persistent. Poverty has resulted in the destruction of the normal lifestyle of theeducated, employed, urban dwellers. In some instances, it has evolved into absolute deprivation (WorldBank 2000 2–3). For society as a whole, poverty has meant the fall of crude birthrate, particularly in theEastern areas of the former Soviet Union, and the decrease of life expectancy in the Russian Federation andthe former Yugoslavia (Table 6.3).

Why have such demographic phenomena become so acute in some regions of the Russian Federation andformer Yugoslavia? Briefly, the

Table 6.3 Sub-regional trends in social welfare (%, change 1989–96)

Real wage Crude birthrate Life expectancy Infant mortality

Central EasternEurope

−16.4 −23.7 2.3 −32.5

South-Eastern Europe −35.3 −21.3 −1.9 −11.7

Countries of formerYugoslavia

– −6.30 – −45.5

Baltics −5.20 −38.9 −2.5 −0.1

CIS and Western FSUcountries

−50.4 −36.7 −5.6 +1.6

Caucasus −76.0 −37.2 −1.7 −20.3

Central Asia −60.6 −30.0 – −19.7

Sources: Ringold (1999, Table 2.1, p. 14) on the basis of TransMONEE Database, UNICEFICDC (Florence, Italy),World Bank.

NotesUnweighted averages are reported for each region. Life expectancy data is for males, at birth. Infant mortality means infant

mortality rate. In this table only, Central Europe includes the Czech Republic, Hungary, Poland andSlovakia. South-Eastern Europe includes Albania, Bulgaria and Romania. Countries of former Yugoslaviainclude Croatia, Slovenia, Macedonia and FR Yugoslavia. The Baltics include Estonia, Latvia and Lithuania.CIS and Western FSU countries include the Russian Federation, Belarus, Moldova and Ukraine. Caucasusincludes Armenia, Azerbaijan and Georgia. Central Asia includes Kazakhstan, the Kyrgyz Republic,Tajikistan, Turkmenistan and Uzbekistan.

answer lies in the combination of local wars, degenerating health services, the stress associated with thedramatic changes of socioeconomic environment, and persistent unemployment. The same conditions have

DIMITRI A.SOTIROPOULOS 217

probably led to well-known phenomena such as the rise of alcoholism and the rise of death rates amongmales, aged 55–59, in the former Soviet Union (Duffy 1998:39, Figure 2.2).

Region-specific causes of persisting poverty

Poverty did not rise equally in all post-communist societies. Between 1989 and 1993 the percentage of peopleliving in extreme poverty rose by twelve times in Bulgaria, by seven times in the Czech Republic, by fourtimes in Hungary and by almost fifty times in Slovakia (calculated on the basis of UNICEF data, presentedin Elster et al. 1998:221, Table 6.2). As noted above, generally speaking, poverty was owed to output fall,income inequalities and the collapse of the firm-based social protection (World Bank 2000:1, 10, 14; Barr2001a:242–243).

To account for persisting cross-national differences in poverty, Barr (2001a) has offered an explanation,distinguishing among post-communist states who were quick reformers and states who were late reformers.Radical reformers of the socialist system, such as Poland and Estonia, experienced substantial poverty,whereas late, reluctant reformers, such as Russia and Romania, experienced even greater poverty. Thedifference lay in the ability of the former to enter more quickly a phase of strong economic growth (ibid.:244–245). These comparisons refer to differences between neighbouring countries, i.e., to intra-regionaldifferences. It may be useful to open the comparison and to contrast the various regions which used to beunder socialist regimes.

In CIS and the other Western areas of the former Soviet Union, poverty may be attributed to the rapidlygrowing inequality after the transition (World Bank 2000:14–16). Rising inequality, in turn, may beattributed to the collapse of wages in the formal economy and to the accumulation of wage arrears. Thosetrends were related to the closing down or to the restructuring of enterprises after the transition and to theinability to create new jobs. In CIS, the prolonged recession combined with the absence of a safety net afterthe transition and caused widespread poverty. Social spending collapsed, as the government lackedresources to finance social policy. While wider economic and welfare reforms remained incomplete,decision-making passed to the hands of former communist cadres and new entrepreneurs, i.e., the newpower elite which was able to capture political power and to rule almost unchecked by any financial orpolitical institutions.

In Central Eastern Europe, the growth of poverty was abated to some extent. Social expenditureremained at relatively high levels, while the social safety net was preserved to a certain degree. Thisdifference with the countries which emerged out of the former Soviet Union may be due to the fact thatCentral European countries had stronger civil societies, more market-supporting institutions and someinstitutional checks and balances to counter the social costs of the transition (World Bank 2000:13, 15).

In South-Eastern Europe, in turn, the situation was worse than in Central Eastern Europe. In the WesternBalkans, poverty was extended as a result of the wars in Bosnia-Herzegovina (1992–95) and in Kosovo(1999) as well as of the total collapse of the state in Albania (1997). These events brought about thedevastation of local economies, the destruction of personal properties, and large migratory movements awayfrom the scenes of conflict. The conflicts also affected countries which did not take part in the wars (e.g.,FYR Macedonia, which has one of the highest unemployment rates in Eastern Europe).

However, even in the Eastern Balkans (Bulgaria, Romania), where there was no war, the situation was notsignificantly better (Sotiropoulos et al. 2003, on the basis of East Western Institute 2000 data). Between1993 and 2000, the percentage of people who were absolutely poor doubled in Bulgaria (it rose from 17 percent to 33 per cent of the population); while in Romania the corresponding percentage was high both at the

218 SOUTH-EASTERN EUROPE: POVERTY

beginning and at the end of that period (32 per cent in 1993, 30 per cent in 2000. see Table 6.4). How haveEast and South East European countries reacted to the above situation?

Table 6.4 Earning inequality (GINI coefficient) and extent of poverty affecting total population in Bulgaria andRomania (%)

GINI coefficient* Poverty**

1991 1992 1993 1994 1995 1996 1997 1987/1988

1993/1994

2000*

Bulgaria 26.2 n.a 25.1 n.a 27.1 29.1 31.4 2 17 33

Romania 20.4 n.a 22.6 n.a 27.8 30.3 42.2 6 32 30

Sources: *East-Western Institute (2000). **Zamfir (1999:95). Data provided by Ileana Neamtu, The RomanianAcademy, Bucharest.

Poverty alleviation measures

In addition to income inequalities, output fall, and the collapse of socialist-era firms, poverty may also havebeen the result of incoherent social policies, reflecting both the deterioration of the economy and thecontradictory political profiles of governments who succeeded each other in power in the post-communistperiod. In the early 1990s, most Central Eastern European countries had tried to protect their poor in threedifferent ways: first, by establishing social safety nets; second, by fixing a minimum wage; and, third, byresorting to practices of the old socialist regimes, such as price subsidies or company-based social funds(Elster et al. 1998:221–222).

In the late 1990s, with the progressive marketization of their economies, many transition countries cut offsubsidies and introduced a means-tested subsistence benefit or allowance in cash. This policy was oftenadministered locally, by municipal authorities. It was a kind of social assistance that mostly applied to theregistered unemployed people. In many cases it was a discretionary temporary allowance (Duffy 1998:88).Thus, often it was not a typical West European-like statutory minimum cash benefit. Its actual contentvaried from one country to the other. Variation existed in regard to the time of the introduction of socialassistance measures, the eligibility criteria, the financing and the efficiency of the benefit’s delivery, and itsintegration into a wider poverty alleviation scheme. At the turn of the twentieth century, there was bothinterregional and intra-regional variation.

By 1999, in Central Eastern Europe, Poland, the Czech Republic, Slovakia and Slovenia had adoptedvarious social assistance schemes, somewhat reminiscent of a cash benefit, whereas Hungary had not doneso. In detail, since 1990 in Poland there has been a means-tested temporary allowance, financed by the statebudget and offered at the discretion of local authorities. The allowance was granted on the basis of anindexed income threshold and “social dysfunctions”, such as homelessness. (Czepulis-Rutkowska 1999:2,4; Consensus Programme 1999:217). This measure was complemented by a social pension, addressed todisabled old-age people and, since 1996, by a special temporary allowance, with a less stringent incomethreshold. Overall, there was no obligatory minimum income guarantee (MIG).

In the Czech Republic, since 1991–92, there has been a social care benefit, administered nationally on thebasis of social need criteria. The benefit was means-tested, tax-financed, indexed and unlimited in duration(Consensus Programme 1999:216, 218, 220, 222). In Slovakia, since 1998 there has been a social assistancecash benefit, provided periodically to citizens in material need. These were old-age or disabled orunemployed people. However, the general thrust of the Slovak system of social assistance was oriented

DIMITRI A.SOTIROPOULOS 219

towards social services, for instance, social counselling, recreation, educational activities and home care(Krckova 1999:2–3). In Slovenia, since 1992 there has been a social assistance benefit (mostly in cash, butsome times in kind). The benefit was offered to people who were permanently incapable of working or wereover sixty years old (Stropnik 1999:5).

In Hungary since 1993 there have been particular forms of social assistance to low-income pensioners, tothe long-term unemployed, to children in need and to other people in need. However, there was no generalassistance scheme and the existing measures were financed by local authorities, which raised the relevanttaxes (Szalai 1999:3, 5, 15; Consensus Programme 1999:216). There was not even any temporary allowancefor people in dire financial need, but only a “transitory assistance” loan. To sum up, in Central EasternEurope the range of social assistance was quite wide and went from no nation-wide assistance at all tonationally administered cash benefits.

In the Baltics, the situation looked different, in the sense that the relevant schemes were more generous.Let us take two examples. In Estonia, where the word “poor” is not used in any official documents (Leppik1999:3), there has been a subsistence benefit since 1995. The benefit was addressed to people with amonthly income below a certain threshold. It was financed both by national and local sources, it wasunlimited in time and its level was determined by the government (Consensus Programme 1999:216, 218,222). However, the subsistence benefit was administered locally and the claimants must have beenregistered as unemployed (Leppik 1999:10,12).

In Lithuania, since 1990 there has been a means-tested social assistance benefit in cash. It was paid tofamilies which earned income below a certain level. The benefit was based on local taxes and wasadministered by municipalities. Since 1993, there has also been a scheme for extraordinary cash benefits,supplied in cases not covered by the existing legislation and paid as a lump sum. And since 1994 there hasbeen a non-contributory social pension, meant for old-age people who did not have the necessary length ofservice to be included in the social insurance system (Cerniauskas 1999:3, 6, 8, 11). Generally, both theEstonian and the Lithuanian systems of social assistance were decentralized. In that respect they resembledmost Central Eastern European systems. In sum, it seems that on the whole they offered wider coveragethan the latter.

In South-Eastern Europe, in turn, the situation was different. As noted above, because of the wars in formerYugoslavia it is difficult to speak of social protection in the north-west side of the Balkan peninsula.However, there are examples of social assistance schemes in four Balkan countries. In Albania, since 1993,there has been a social assistance scheme (Hutton et al. 2000:128). It included a cash benefit, provided bythe state budget, but administered by local government. It was means-tested and was complemented by foodaid and lump-sum payments in cases of special need (Shalari 1999:2–3, 17).

In Bulgaria, since 1998 there has been a social assistance package which included a monthly benefit toindividuals or families whose income in the previous month fell below a certain level. This was a temporary,means-tested measure. Additional cash benefits were granted in some instances, e.g., to pay the rent ofpublicly-owned homes and for the heating of houses during the winter (Stoyanova and Kirova 1999:4–6).Benefits in kind were also envisaged, while in some areas the non-payment of electricity bills was tolerated(Sotiropoulos et al. 2003). Also, for particular categories of old-age people, there was a social pension.

In FYR Macedonia, since 1997 there has been a scheme providing constant cash benefit to individualsunable to work and, since 1998, another scheme for the working poor, i.e., for employed people orpensioners who were paid very lowly, and for the unemployed. The benefit was means-tested, could be grantedfor a total of four years and was administered by local “centres for social work” (Nikolova 1999:4–5, 10, 22).Overall, however, Macedonia was particularly hit by unemployment. In 1997, unemployment was around30 per cent, but only one out of three unemployed people received unemployment benefits. Most of the

220 SOUTH-EASTERN EUROPE: POVERTY

unemployed did not receive benefits because they had been without a job for more than eighteen months(Hutton et al. 2000:134).

Finally, in Romania, since 1995 there has been a social aid scheme, which included a means-testedmonthly cash benefit, available to the registered unemployed. It was mostly funded and administered bylocal authorities. In cases of extreme need, there was an emergency lump-sum benefit. In Romania, as inother South-East European countries, it proved that the amount was insufficient, the bureaucraticinefficiencies were frequent and the loopholes of the relevant laws were also quite significant. In 1998 inRomania a special fund to alleviate poverty was created by law. In 1999, another social assistance law waspassed (Racoceanu 1999:6, 8, 20–21, 36–37). However, as we shall see, it was not until very recently that acomprehensive anti-poverty policy was put together. In sum, social assistance schemes in the four South-East European countries surveyed above looked problematic. The usual problems of fiscal stringency, over-centralization, inadequate legal framework and administrative incapacity have compromised the socialassistance schemes in place.

The general picture coming out of the above variety of schemes conforms to what Standing had observedin the mid-1990s:

Since the outset of the reform process, governments in most countries tried to reform their system ofsocial protection by shifting from the patchwork of universalistic, mainly employment-related schemesto greater reliance on social insurance geared to cover market-related “risks” or contingencies and tomeans-tested social assistance. They also resorted to greater “targeting” of benefits, which meantadopting a new array of conditions for entitlement to old and to new forms of benefit.

(1996:236)

The question is, how did the new schemes perform?

Evaluating the poverty alleviation efforts

The available evidence as regards the coverage, efficiency (i.e., share of transfers to the bottom incomedecile) and effectiveness (i.e., social assistance income benefits as share of the recipients’ totalexpenditures) refers to a few transition countries and refers mostly to the first half of the 1990s (World Bank2000:296–297, Tables 9.1, 9.2, 9.3).

Summarizing the results of household surveys carried out by Milanovic in Poland, Estonia, Hungary,Russia and Bulgaria in 1993–95, Barr (2001b: 184–185), concludes two things: first, in the early 1990s thepercentage of total assistance expenditure going to the poorest 10 per cent of households was low. Even inthe best case, which is Estonia, this figure was only 34.7 per cent, while in the worst case, which was theRussian Federation, it was 8.2 per cent. So the social assistance schemes were not efficient. Second, theywere not effective either: in 1993 Hungary, which was the best case, assistance relieved less than 30 per centof the poverty gap of the lowest income docile. The corresponding figure for Bulgaria in 1995 was only 1.3per cent (ibid.: Table 10.4).

Another problem is the “growing regionalization of social policy” (Standing 1996:250). In the largercountries of Eastern Europe, such as the Russian Federation, Poland and Ukraine, poverty varies a lotamong regions as does the quality of services offered at the local level. Already in the mid-1990s intra-national disparities of poverty had become extreme. Today, in the Russian Federation poor areas can hardlyafford to protect the poor, while other areas (e.g., the Moscow city region) have proved to be comparativelygenerous (Alexandrovna and Braithwaite 2000:230). Other cases of regional and local variation in poverty

DIMITRI A.SOTIROPOULOS 221

alleviation are common in Eastern Europe (e.g, Hungary, Estonia, Lithuania; see above). Assistanceschemes have not been successful in addressing the problem of regional disparities either.

An additional problem is that a usual measure of welfare scheme evaluation, i.e., the percentage of thepoor covered by the relevant poverty alleviation measures, may not be useful. This measure says somethingmore about the narrowness of—country-by-country—official definitions of poverty than about the results ofthe fight against poverty. At any rate, in Bulgaria, Hungary and the Russian Federation, many non-poorwere included in the poverty alleviation schemes. Conversely, it seems that in Poland, Bulgaria, the RussianFederation and Latvia, the majority of the poor were excluded from the relevant schemes. The World Bankhas concluded that “in most countries, social assistance programs do not do a particularly good job ofreaching the poor, concentrating their expenditures on the poor, and lifting the poor out of poverty” (2000:299). A closer look at two South-East European countries may illustrate our points about policy priorities,incoherence and inefficiency.

Poverty and social assistance in Bulgaria and Romania

In Bulgaria and Romania, income, social insurance and public health policy shifts after the transition had anegative impact on the living standards of large segments of the population. In other words, policy shiftsadversely affected income inequality, which became acute, and poverty, which increased in the 1990s(GVG-Bulgaria 2002; GVG-Romania 2002).

In both Bulgaria and Romania, after 1989 there was the rise of extreme or severe poverty (Todorova1999; Rotariu and Popescu 1999; Molnar 2000; Mitev et al. 2001; Magyari et al. 2001). Pockets of povertyemerged and persisted in various regions. Deprivation also had some other aspects, such as the persistenceof long-term unemployment among vulnerable groups (low-skill workers, members of minorities) and thehigh institutionalization rate of children who lived in caring homes.

In Bulgaria, as we noted above, between 1993 and 1997 there was an increase of income inequality andan increase occurred in poverty. According to a Bulgarian source, in the first half of the 1990s thepercentage of households the members of which individually had income lower than 50 per cent of theaverage Bulgarian individual income, ranged between 8 per cent (in 1991) and 15 per cent (in 1994) of allhouseholds (Shopov 1999:7l).

Objective conditions of poverty were also reflected at the level of attitudes: in 2002 in Bulgaria, 85 percent of the population considered themselves to be poor. (However, there was no official poverty line.)Independent researchers, who published a report in early 2002 (Bulgarian newspaper 24 Hours, 20 March2002), set the poverty line at the equivalent of � 49 per month. To put this figure in context, we may cite thefollowing levels of monthly salaries (drawn on press reports of the Bulgarian National Radio): in early 2002in Bulgaria, the average monthly salary of a doctor working in the public health system was theapproximate equivalent of � 160, of a coal-worker or miner � 175–200, of a primary school teacher � 110, ofa civil servant � 150 and of a worker working at a sewing machine � 50 (but parliamentary deputies earnedapproximately � 650 per month).

As noted above, in the late 1990s, social assistance in Bulgaria consisted of cash benefits, provision ofessential goods (for example, coal) and de facto subsidies (for example, turning a blind eye to arrears inpaying electricity bills). This ad hoc approach is gradually being replaced by a more liberal, residualistapproach. In February 2002 a new law for the social rehabilitation and integration of disabled people waspassed and in March of the same year a law to assist poor children was adopted. The latter law wouldincrease the monthly assistance to poor children from the equivalent of � 4.26 to � 7.5. The same law

222 SOUTH-EASTERN EUROPE: POVERTY

stipulated that eligible were only the children belonging to families in which each member had a monthlyincome of less than � 75.

In Romania, since 1995 the inequality gap has become wider than in Bulgaria. Between 1993 and 2000,the share of the poor in the total population remained steadily high, at about 30 per cent (Table 6.4). Thismay be explained by the following fact: already since the beginning of the transition, the level of publicspending on social protection in Romania was much lower, not only than that of Bulgaria, but also lowerthan the equivalent OECD levels and the levels of other East European countries. This level did not rise fasteither. For instance, in 1992–93 in Romania the level of social expenditure rose only by 1 per cent, while inother transition countries it rose by 6 per cent.

Throughout the transition period, the social assistance policy of Romania was at best incoherent. Publicexpenditure for social assistance ranged between 0.1 per cent of GDP (in 1994) and 0.4 per cent (the highestpoint, early in the transition, in 1991). There was no data on the actual number of assistance beneficiaries.There were cash benefits, but they probably reached a small number of the needy people, while means-tested assistance was introduced only in October 1995 (Zamfir 1999:92; Zamfir 2000:25).

Recently, a more coherent attitude has been adopted by the government of Romania, which openlyrecognized the problem of severe poverty. It adopted a National Action Plan to fight poverty and socialexclusion (similar to the “NAPincl” plans implemented by EU member states) and set up a specialcommittee (the CASPIS committee) to monitor the situation. In July 2000, a National Child Welfare SystemReform Strategy was adopted and in July 2001 a minimum income guarantee law was passed.

To sum up this section, in the area of social assistance, the governments of Bulgaria and Romania willhave to deal with the most pressing problems of all welfare problems (GVG-Romania 2002:114–117;Government of Romania—CASPIS 2002, GVG-Bulgaria 2002:108–114). In Bulgaria the main objectives,set by the government, are to re-distribute competences between central and local authorities; to monitorinstances of poverty much more closely so as to apply a differentiated approach to means-tested schemes; tocontinue providing non-income support, including support to socially excluded families, vocationaleducation and in-kind benefits; and to further engage non-governmental organizations (NGOs) in socialassistance. In Romania, on the other hand, the government has also set several objectives. These include theeradication of severe poverty and social exclusion, the bridging of the income gap between pensioners andmembers of the labour force, the promotion of social cohesion and the improvement of the welfare ofchildren. Compared to even the recent past, this looks like a re-arrangement of policy priorities and a morecoherent strategy for the future.

Strategies for the future: lessons from Southern Europe?

However, today, more than ten years into the transition, many post-communist states still have an ad hocapproach to emerging social problems (Deacon 2000:149). This approach includes appeals to philanthropyand voluntary organizations and transfer of the control over social assistance schemes to local governments,which themselves may be impoverished and very discretionary in the way they hand out benefits to thepoor.

Ferge (2001:20–21) claims that the region is gradually adopting an orthodox neoliberal agenda of welfarereform. This may be true for the emerging three-pillar systems of pensions in a few Central EasternEuropean states (Bastian 1998), even though, in regard to pensions, there is also an alternative, lessorthodox liberal four-pillar model (Wagener 2002: 165).

The social assistance schemes, on the other hand, are not necessarily becoming neo-liberal. They are ofthree different types (based on the work of Milanovic, cited in Barr 2001b:184): first, “concentrated”

DIMITRI A.SOTIROPOULOS 223

schemes, which generously disburse to few recipients (e.g., Poland, Estonia); second, “dispersed” systems,which disburse small benefits to many recipients (e.g., Hungary); and third, “irrelevant” schemes, whichdisburse small benefits to few recipients (e.g., Bulgaria). So there is a variety of schemes currently in place,which seem to be still in the making, thus reflecting the larger state of flux in which East and South-EastEuropean welfare states are found. As Bob Deacon has put it:

For the countries of Central Europe, and some of the Balkan and Baltic ones, the question of whethertheir overall social policies would gradually evolve towards a Western European, conservative,corporatist model, or towards a US liberal kind, was generally undecided by the late 1990s.

(2000:152)

The state of flux, which we also noted in the beginning of this chapter, is probably due to the disintegrationof the earlier paternalistic, statist welfare regimes and to the gradual building of a new kind of statemechanism. This mechanism reflects the rise of a new model, the model of the regulatory state, which hasalready been spotted both at the national level (Majone 1997) and at the level of the European Union (e.g.,Walby 1999). The regulatory state taxes less and spends less, but may in some sense intervene more ineconomy and society through a multitude of legal frameworks and mechanisms, supervising economicactivity.

If we have in our picture a fading paternalist state and an emerging regulatory state, what alternativestrategies are available in the fight against poverty? We will briefly look at what the EU has to offer and atthe imitable strategies of Southern European EU member states and then summarize the stance of the WorldBank on this issue.

The possible impact of accession to the European Union

The EU is primarily concerned with the macro-economic stability of its member states and the states ofCentral Eastern Europe and the Baltics which have recentiy become full members of the EU. The EU doesnot have a single welfare model. By contrast, as is widely known, EU member states have for a long timenow adopted different models of social protection or combinations thereof. This is no longer a point of fact,but has acquired a normative status, owing to the Treaty of Amsterdam. The Treaty provides for anunnamed variety of national preferences as far as social security is concerned. In that sense, new incomingmember-states have more degrees of freedom in regard to their social policies than their macro-economicpolicies (Wagener 2002:152, 153, 162). Obviously the two kinds of policy are closely related. In addition,the EU is not “neutral” when it comes to labour policy which by definition is part and parcel of a widelyconceived welfare policy. For instance, the EU favours flexible labour markets. However, it has offered—toold and new member states alike—little specific guidance as to structures, pro grammes and levels ofwelfare provision, a tendency which has been particularly evident in the area of anti-poverty measures(Leibfried 1992: 255–256).

On the other hand, in contemporary Europe, the context of such measures is neither completelyundetermined nor totally open to the preferences of individual member-states. There is a diffuse Europeanset of values to which anti-poverty measures may conform. The values include social solidarity, somesubsidiary state support in times of economic hardship and some kind of nation-wide health coverage.These principles are valuable, but also too abstract to serve as models for safety net construction,particularly for the new, EU-member states.

224 SOUTH-EASTERN EUROPE: POVERTY

The paradox is that upon being “translated” into concrete policy measures, the above abstract values maysidetrack East European countries into welfare regimes which differ from the traditional European ones.The EU, through its criteria for enlargement, requires large economic adjustments and also favours theprivatization of education and health systems in Eastern Europe. This latter tendency points to a newwelfare system tailored to the transition countries. It is different from that of the traditional West Europeanwelfare system and may well lead to an increasing gap between Eastern and Western Europe (Ferge 2001:13, 20, 22). Pascall and Manning suggest that in Eastern Europe there is no political dynamic pro viding forthe formulation of social policies which would bridge the gap between East and West. In fact, nationaleconomic inclusion of the transition states into the EU as new member states may go hand in hand withwidening social exclusion in the transition societies (2002:152).

However, even these latter authors, who see the future of East European welfare as bleak, perceive somepossibilities for improvement. As a general strategy, they consider that, as was the case with the countriesof Southern Europe in the 1980s, the poorer East European countries may achieve particular treatment fromthe EU. For instance, they may be able to claim the creation of a new cohesion fund aimed at theirparticular needs, in exchange for meeting the monetary and other economic criteria of the Union’s enlargement(ibid.: 156). Such a deal may be an interesting, but very uncertain strategy for the future. The deal dependson the will of the most powerful of the two partners of the bargain, which is the EU. If attained, the dealwould nevertheless not solve the problems of planning and implementation of social assistance schemes,which are internal to the East European systems. As regards Southern Europe, it is difficult to show that thevarious “packages” (including the “Delors packages”) offered by the EU to Greece, Spain and Portugal, inthe first fifteen years of their EU membership, had any significant effect on curbing poverty in thesecountries. Rather, it is the experience of individual Southern European states, to which we now turn, thatmay prove more useful to East European reformers.

The Southern European experience and its relevance to Eastern Europe

Even though there are obvious differences between Southern European and East European welfare states, itmay be useful to look for specific comparable experiences with constructing safety nets in Southern Europe(Matsaganis et al. 2003). After all, some of the problems of social protection in Eastern and South-EasternEurope have already been observed in Southern Europe. In the latter region in the past there were“rudimentary welfare regimes” as far as poverty alleviation is concerned (Leibfried 1992: 253–254). Todaythere is a more developed “Southern European model” of the welfare state (Ferrera 1996), although it doesnot fit Greece, Italy, Portugal and Spain equally well.

Certain features found both in some of these four Southern European countries and in post-1989transition countries are the following:

1 Comparatively large agricultural populations, which means that there may be a urban-rural divide, withsome rural areas experiencing extended poverty.

2 Widespread informal economic activity and concomitant tax evasion, which is translated, first, intoskewed and often unknown income distribution and, second, into the inclusion of non-poor in povertyalleviation schemes.

3 Relatively extended and traditional households, which means that part, if not most, of the care for poor,unemployed or disabled people is assumed by female family members. This situation, which isencountered more in South-Eastern Europe than in Eastern Europe in general, may decrease pressure onpaternalist states to weave budgetfunded safety nets.

DIMITRI A.SOTIROPOULOS 225

4 Limited administrative capacity, which indicates an inability to fully implement welfare reform and itsprerequisites, such as the rise of the state’s tax-raising capacity. An increase in general administrativecapacity would be most useful in post-communist societies, in order to rescue the state mechanism frompowerful private interests.

The differences in the level of development, structure of the economy, and culture between SouthernEurope, on the one hand, and Eastern and South-East Europe, on the other, are far from negligible.However, some enduring problems of Southern Europe and, also, attempts to alleviate poverty in that regionmay contribute relevant input to the formulation of relevant East European policies. In terms of enduringproblems, take, for instance, the case of the “perforated” safety net, found in Greece (Matsaganis 2002:2,and Chapter 2 in this volume). This type of net is full of holes and does not prevent the descent ofconsiderable numbers of people into poverty. In Greece there is no minimum income guarantee (MIG)scheme. This is a problem related to the general structure of the Greek welfare state which heavily leanstowards the social protection of pensioners. The employees and pensioners of well-endowed public sectorsocial security fund are comparatively well protected. These are the “insiders” of the system, who have putup very strong resistance to any meaningful reform of the Greek welfare state.

The case of Portugal is different but also telling. In that country an interesting new safety net has beenput in place since 1996. It involves, among other things, a MIG scheme plus a contract signed between therecipient of the benefit and the local government agency in charge of dispensing benefits. The contract aimsat the re-integration of the typical, socially excluded recipient into the society. This is an individualizedapproach which involves social workers assessing the real need and performance of petitioners of thebenefit. Despite its success, the pro gramme has nevertheless encountered problems of implementation, suchas fraud in access to the benefit (Capucha et al. 2002:20–21, 23, 37 and Chapter 5 in this volume).

How could the Portuguese and Greek experiences be relevant to post-communist societies? Similarproblems of perforated nets and over-protected social categories can be found in Eastern and South-EasternEuropean countries. In those countries there is considerable misallocation of social assistance in terms ofrecipients and allocated funds (Braithwaite et al. 2000:165, Table 4.2). There is also inertia and resistance towelfare state reform in certain transition countries. Resistance may come, first, from groups of manualworkers and public sector employees who used to be well protected under the socialist system and, second,from employees of larger enterprises which have put together specific schemes for their personnel. Owingto fragmentation of benefits in the post-communist era, some groups may stand to lose the access whichthey enjoy to enterprise-based benefits or new social insurance and private benefits (Standing 1996:232).

More generally, even though external economic transformations cause problems to national welfaresystems, welfare reform is inhibited more by national institutional logics and internal dynamics (Ferrera1997:10–14). At least for West European welfare systems, “the principal sources of… challenge areinternal” (Ferrera and Rhodes 2000:3). In other words, the risks brought about by globalization may be lessonerous than it is often presumed, even for such open and weak economies as those of Eastern Europe.

Another problem which is common to contemporary South-East European and East European welfarestates is regional disparities in treating the poor. For example, the evaluation of the RMI, the Italian MIGscheme, has shown extended territorial disparities, including different eligibility thresholds and hugeinequalities among municipalities in terms of capacity to administer the benefit (Sacchi et al. 2002:2, 27,31, 35 and Chapter 3 in this volume). As we have noted, similar but perhaps more pronounced regionalinequalities have been found in countries making the transition from socialism. Such and other problems arerelated to fiscal constraints, institutional incapacities and demographic trends showing a gradual increase inthe share of elderly people in the total population (Bastian 1998:69).

226 SOUTH-EASTERN EUROPE: POVERTY

The World Bank’s view

All the above trends have led commentators of East European welfare systems to pessimistic conclusions.Ferge claims that a major international institution, the World Bank, which exercises influence in the socialpolicies of East and South-East European states, is mainly interested in the privatization of pensions (2001:21). Wagener says that the Bank is interested in the accumulation of pension funds in private insurancecompanies with an aim to enhance economic growth (2002:163).

Indeed, the World Bank offers a strategy for the future of welfare in transition countries. The cornerstoneof the World Bank’s thinking is the promotion of economic growth. “Provided it is equally shared, growthcan have a big impact on poverty reduction” (World Bank 2000:19). A relevant idea is that anti-povertypolicies could offer temporary protection to the disadvantaged segment of the workforce, so that thissegment is trained better in order to fit the new economic environment. In this view, the hope is that “aslong as the economy on the whole is growing some optimism about a growing stock of human capitalalongside the stock of physical capital is justified” (Hoelscher 1999:171).

The former idea (of the World Bank) is more sensitive than the latter to the following well-knownproblem: to whom will the returns of growth accrue? If economic growth in post-communist societies is tobe “equally shared” (as the World Bank states), then the traditional trade-off between efficiency and equitymay be conceived in a different manner, i.e., it should not be regarded by East European governments as azero-sum game any longer. Two considerations may help us understand why. First, a one-dimensionalpolicy emphasis on stabilization and rapid growth at all costs will sooner or later breed the illegitimacy ofthe political regime and bring destabilization of the political system as a whole, putting whatever growthhas been attained to the strains of another regime crisis. Second, in terms of popular expectations, if early onthere are convincing signs that governments care to construct a social safety net alongside promotingeconomic growth, then “chances will be greater that a majority of the population will accept the risk ofeconomic reforms” (Scholz and Tomann 1999:96) and increase their productivity. In these respects, equityand efficiency may be recognized as complementary rather than competing principles.

However, as the World Bank itself admits, there may be a considerable time lag betweenaccomplishment of growth targets and job creation, let alone income re-distribution. The latter is veryuncertain, if indeed the state in Eastern and South-Eastern Europe has been “captured” by private interests.If the economic strategy suggested by the World Bank is one-dimensional (i.e., economic growth), itspolitical strategy is multifold. It involves increasing political accountability, strengthening institutionalchecks on powerful private agents, political elites and the mass media, fostering civil society participationand private sector competition, and, of course, reforming public administration (World Bank 2000:15–17,179–180).

The strategies suggested by the World Bank are not the same for all transition countries: higher-income,transition economies may want to proceed with pension reform, more privatizations and refined regulatoryframeworks. Lower-income, transition economies may want to impose harder budget constraints, enablesmall business development and agricultural growth and proceed with land reform. In either case, poverty willbe transmitted to the next generations unless, in addition to social assistance, care is taken to providetoday’s poor and their children with highquality health care and with educational skills (ibid.: 20–21,181). Obviously, all this requires an improvement in the fiscal situation, which, in turn, cannot be effectedwithout improving tax compliance, on the side of state revenues, and targeting and effectiveness of cashbenefits to the poor, on the side of state expenditures.

There is a clear need to reshape the East European institutional frameworks for formulating,implementing and evaluating social safety nets (Standing 1996:249) and to increase their relevantadministrative capacity (Barr 2001b:181). Which brings us full circle to the problem of reform of the state

DIMITRI A.SOTIROPOULOS 227

apparatus, which we have hinted at in the beginning of this chapter. In Eastern and South-East Europe,welfare reform and administrative reform go together and the latter is a prerequisite of the former.

Epilogue: who could be the reformers?

In this chapter we have argued that poverty in post-communist Eastern and South-East Europe was theresult not only of the transition to the market system, but also of the new social class cleavages, the politicalpriorities of incoming governments and the incoherent social assistance policies. The average citizen hasbecome a poor person. In terms of cross-national variation, the worst cases of income inequality and povertyare the CIS and the South-East European countries, while the Central East European countries have faredbetter. Poverty alleviation measures were often means-tested and targeted. They have proved to beineffective and inefficient. Despite the acuteness of such problems, most welfare reform has concentratedaround the reform of pensions. Very recently in some countries, such as Bulgaria and Romania, there hasbeen a more concerted effort to formulate social safety nets. General strategies for the future include anemphasis on higher economic growth and multiplepillar social insurance systems. More specific strategiesavailable to East European countries may benefit from the experience of the Southern members of the EU toput together safety nets in the 1990s.

Given the problems of implementation of social safety nets, the enduring poverty in Eastern Europe as awhole, and the regional disparities among regions within many East European countries, a main prerequisitefor any welfare reform is to proceed with the reform of state institutions. Both welfare reform andadministrative modernization are difficult to attain. To succeed, the aim of prospective reformers would beto increase the cost of “capturing” the state for business elites and also the cost of using it for narrow privateinterests. But who could be these reformers?

Writing about the problem of the agency of social policy reform in the first phase of the transition, ClausOffe has noted the “associational wasteland” and the “amorphous social-structure” of post-communistsocieties (1996:240, 242). He also mentioned the conditions for the emergence of “associability” which inturn is a prerequisite for a “society’s capacity to develop a social policy system (ibid.: 1996:241).Conditions included, first, the rise of the people’s expectations from increased participation in associationsand, second, the emergence of clear social cleavages, which, however, would “be amenable to compromise”(ibid.).

Such pressure “from below” is indispensable for social policy reform. However, the spread of ideas andthe forging of suitable political alliances at “the top” are also needed. Who could be the catalysts of change,able to influence political elites? They may be civil society associations, modernizing policy-makers andreform-minded intellectuals, prepared to stand up and speak out in favour of socially excluded interestgroups. Under the circumstances, it is clearly difficult to build a reform-minded coalition which would gainhegemony in the social policy community of post-communist societies. Yet, reformers stand a chance ofsuccess, if they work in the context of converging pressures from international organizations and in theprospect of joining the EU, a prospect that has caught the imagination of the East European peoples. AsEast European countries are currently forging a variety of relations with the EU, a fuller analysis of thepolitics of social reform would have to develop an “advocacy coalition framework” (Sabatier 1998), on aregion- and country-specific basis.

To conclude, general recipes of state reform will not do. Obviously, there are administrative problemscommon to all post-communist states. These include corruption, lack of training in modern skills andmanagement methods and little use of new technologies. However, such problems differ a lot from oneregion of Eastern Europe to the other and from one country to another. As a result, the ability of involved

228 SOUTH-EASTERN EUROPE: POVERTY

states to deliver welfare services varies a lot. This is a typical conclusion of all relevant studies (e.g.,Braithwaite et al. 2000:91, 158).

An obvious consequence of this conclusion is the need to avoid wholesale strategies for the future and tostart a region-by-region and country-by-country analysis of possible future alternatives. A secondconsequence is that, given the complexity and variability of the problem under study, a single, isolatedpolicy measure will not do. For instance, by itself, the introduction or improvement of MIG schemes in theEast and South-East European countries is not a panacea. In other words, MIG, which should be availableas a right of citizenship, is a necessary but not sufficient condition. In Eastern Europe, MIG should becomplemented by specific transfers made in cases of need and by a net of local social services (Standing1996:252). A relevant interesting lesson may be learned from the Italian case. Writing about the effects ofintroducing MIG in the context of poverty-stricken areas, Ferrera concludes that in local contexts which arecharacterized by mass poverty due to economic stagnation, the right policy is a mix of instruments aimed atpromoting economic development, subsidizing/activating the unemployed, supporting families as such —with the RMI (the minimum income guarantee) serving as a net for the tails (Ferrera, Chapter 1 in thisvolume).

Finally, in each national case, the right social policy mix by itself will not suffice, unless it isaccompanied by measures to create growth, to curb unemployment, and to improve the health and educationsystems. In a word, social assistance policy may work, if it is put in the context of wider efforts to overcomeoverall economic and social stagnation.

Acknowledgements

The author would particularly like to thank Professors Maurizio Ferrera (Milano), Luis Capucha (Lisbon),Manos Matsaganis (Athens), Jens Bastian (Athens) and Nick Manning (Nottingham) for information,encouragement, and feedback. He acknowledges the invaluable research assistance of Ileana Neamtu(Bucharest) and Maya Stoyanova (Sofia).

The section on pp. 282–284 was written with the help of Ileana Neamtu and Maya Stoyanova.

References

Alexandrovna, A. and Braithwaite, J. (2000) “Targeting poverty benefits in Russia: reality-based alternatives to income-testing”, in S.Hutton and G.Redmond (eds), Poverty in Transition Economies, London: Routledge, pp. 228–247.

Atal, Y. (1999) “Introduction”, in Y.Atal (ed.), Poverty in Transition and Transition in Poverty: Recent Developmentsin Hungary, Bulgaria, Romania, Georgia, Russia, Mongolia, Oxford: Berghahn and Paris: UNESCO, pp. 1–31.

Barr, N. (2001a) The Welfare State as Piggy Bank: Information, Risk, Uncertainty, and the Role of the State, Oxford:Oxford University Press.

Barr, N. (2001b) “Reforming welfare states in post-communist countries”, in L.T. Olrowski (ed.), Transition andGrowth in Post-Communist Countries: The Ten-Year Experience, Cheltenham: Edward Elgar, pp. 169–218.

Bastian, J. (1998) “Introducing private pension funds in transition economies of Central Europe”, in J.Bastian (ed.), ThePolitical Economy of Transition in Central and Eastern Europe: The Light(s) at the End of the Tunnel Aldershot:Ashgate, pp. 65–93.

Bauman, Z. (1993) “After the patronage state: a model in search of class interests”, in C.G.A.Bryant and E.Mokryzcki(eds), The New Great Transformation? Change and Continuity in East-Central Europe, London: Routledge,pp. 14–35.

Braithwaite, J., Grootaert, C. and Milanovic, B. (2000) Poverty and Social Assistance in Transition Countries, London:Macmillan.

DIMITRI A.SOTIROPOULOS 229

Capucha, L., Bomba, T., Fernandes, R. and Matos, G. (2002) “Fighting poverty and social exclusion in SouthernEurope: the case of Portugal”, paper presented at the international seminar on “Fighting Poverty and SocialExclusion in South-ern Europe: Dilemmas of Organization and Implementation”, Bocconi University, ISPI, Milan,Italy.

CASPIS (2002) National Anti-Poverty and Social Inclusion Plan (The CASPIS Commission Report), Bucharest:Government of Romania.

Cerniauskas, G. (1999) “Lithuania”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Consensus Programme (1999) Mutual Information System on Social Protection in the Central and Eastern EuropeanCountries: Czech Republic, Estonia, Hungary, Poland, Slovenia, PHARE, Brussels: European Commission.

Cox, T. and Mason, B. (1999) Social and Economic Transformation in East Central Europe: Institutions, PropertyRelations and Social Interests, Cheltenham: Edward Elgar.

Czepulis-Rutkowska, Z. (1999) “Poland”, in Consensus Programme, Sector: social assistance, national survey, finalversion, PHARE, European Commission, ADECRI/PARTEX (July).

Deacon, B. (ed.) (1992) The New Eastern Europe: Social Policy. Past, Present and Future, London: Sage.Deacon, B. (1997) “The making of post-communist social policy: the role of international agencies”, The Journal of

Social Policy, 26(1):43–62.Deacon, B. (2000) “Eastern European welfare states: the impact of the politics of globalization”, Journal of European

Social Policy, 10(2):146–161.Dufify, C. (1998) Opportunity and Risk: Trends in Social Exclusion in Europe, Project on Human Dignity and Social

Exclusion (HDSE), Brussels: Council of Europe (Aprill 998).East-Western Institute (2000) “Task force on economic strategy for South Eastern Europe”, New York-Prague-Moscow-

Kyiv-Brussels-Helsinki: East-Western Institute.Elster, J., Offe, C. and Preuss, U.K. (1998) Institutional Design in Post-Communist Societies: Rebuilding the Ship at

Sea, Cambridge: Cambridge University Press.Emigh, R.J., Fodor, E. and Szelenyi, I. (2001) “The racialization and feminization of poverty?” in R.J.Emigh, E.Fodor

and I.Szelenyi (eds), Poverty, Ethnicity, and Gender in Eastern Europe during the Market Transition, Westport,CT: Praeger, pp. 1–32.

Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism, Princeton, NJ: Princeton University Press.Eyal, G., Szelenyi, I. and Townsley, E. (1998) Making Capitalism Without Capitalists: The New Ruling Elites in

Eastern Europe, London: Verso.Ferge, Z. (1992) “Social policy regimes and social structure: hypotheses about the prospects of social policy in Central-

Eastern Europe”, in Z.Ferge and J.-E. Kolberg (eds), Social Policy in a Changing Europe, Frankfurt: Campus andBoulder, CO: Westview Press, pp. 201–222.

Ferge, Z. (1997) “The changed welfare paradigm: the individualization of the social”, Social Policy and Administration,31 (1):20–44.

Ferge, Z. (2001) “European integration and the reform of social security in the accession countries”, The EuropeanJournal of Social Quality, 3 (1):9–25.

Ferrera, M. (1996) “The southern model of welfare in social Europe”, Journal of European Social Policy, 6 (1):17–37.Ferrera, M. (1997) “Welfare state reform as an institutional challenge”, unpublished paper presented in the workshop

“Political Theory and Social Policy”, Maison des Sciences de l’Homme, Paris, May.Ferrera, M. and Rhodes, M. (2000) “Recasting European welfare states: an introduction”, West European Politics,

Special Issue on Recasting European Welfare States, 23 (2):1–10.GVG-Bulgaria (2002) Study of the Social Protection Systems in the 13 Applicant Countries —Romania Country Report,

Second Draft, Cologne—Brussels (October).GVG-Romania (2002) Study of the Social Protection Systems in the 13 Applicant Countries —Romania Country

Report, Second Draft, Cologne—Brussels (October).Hoelscher, J. (1999) “Social cohesion and transition dynamics”, in I.Collier, H. Roggerman, O.Scholz, and H.Tomann

(eds), Welfare States: East and West, Basingstoke: Macmillan, pp. 167–173.

230 SOUTH-EASTERN EUROPE: POVERTY

Hutton, S. and Redmond, G. (2000) “Poverty in transition economies: an introduction to the issues”, in S.Hutton andG.Redmond (eds), Poverty in Transition Economies, London: Routledge, pp. 1–13.

Hutton, S., Kolpeja, V., Nokvoska, B., Sakiqi, S. and Stoevska, V. (2000) “Albania and Macedonia: transitions andpoverty”, in S.Hutton and G.Redmond (eds), Poverty in Transition Economies, London: Routledge, pp. 125–139.

Krckova, Z. (1999) “Slovakia”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Leibfried, S. (1992) “Towards a European welfare state? Integrating poverty regimes into the European Community”, inZ.Ferge and J.-E.Kolberg (eds), Social Policy in a Changing Europe, Frankfurt: Campus and Boulder, CO:Westview Press, pp. 245–279.

Leppik, L. (1999) “Estonia”, in Consensus Programme, Sector: social assistance, national survey, final version, PHARE,European Commission, ADECRI/ PARTEX (July).

Lipsmeyer, C.S. (2000) “Reading between the welfare lines: politics and policy structure in post-Communist Europe”,Europe-Asia Studies, 52(7):1191–1211.

Magyari, N.L., Magyari-Vincze, E., Popescu, L. and Rotariu, T. (2001) “The social construction of Romanian poverty:the impact of ethnic and gender distinctions”, in R.J.Emigh, E.Fodor and I.Szelenyi (eds), Poverty, Ethnicity, andGender in Eastern Europe during the Market Transition, Westport, CT: Praeger, pp. 123–155.

Majone, G. (1997) From the Positive to the Regulatory State: Causes and Consequences of Changes in the Mode ofGovernance, Working Paper no. 1997/93, Madrid: Instituto Juan March de Estudios e Investigaciones.

Matsaganis, M. (2002) “Fighting poverty and social exclusion in Southern Europe: the case of Greece”, paper presentedat the international seminar on “Fighting Poverty and Social Exclusion in Southern Europe: Dilemmas ofOrganization and Implementation”, Bocconi University, ISPI, Milan, Italy.

Matsaganis, M., Ferrera, M., Capucha, L. and Moreno, L. (2003) “Mending nets in the South: anti-poverty policies inGreece, Italy, Portugal and Spain”, Social Policy and Administration, 37 (6): forthcoming.

Milanovic, B. (1998) Income, Inequality and Poverty During the Transition from Planned to Market Economy,Washington, DC: The World Bank.

Mitev, P.-E., Tomova, I. and Konstantinova, L. (2001) “The price of procrastination? The social costs of delayedmarket transition in Bulgaria”, in R.J.Emigh, E. Fodor and I.Szelenyi (eds), Poverty, Ethnicity, and Gender inEastern Europe during the Market Transition, Westport, CT: Praeger, pp. 33–66.

Molnar, M. (2000) “Poverty measurement and-income support in Romania”, in S. Hutton and G.Redmond (eds),Poverty in Transition Economies, London: Routledge, pp. 109–124.

Nikolova, Z. (1999) “FYROM”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Offe, C. (1996) Modernity and the State: East, West, Cambridge: Polity Press.Pascall, G. and Manning, N. (2000) “Gender and social policy: comparing welfare states in Central and Eastern Europe

and the former Soviet Union”, Journal of European Social Policy, 10 (3):240–266.Pascall, G. and Manning, N. (2002) “Social Europe East and West?” in M.Ingham and H.Ingham (eds), European

Union Expansion to the East: Prospects and Problems, Cheltenham: Edward Elgar, pp. 134–161.Racoceanu, N. (1999) “Romania”, in Consensus Programme, Sector: social assistance, national survey, final version,

PHARE, European Commission, ADECRI/PARTEX (July).Ringold, D. (1999) “Social policy in postcommunist Europe: legacies and transition”, in L.J.Cook, M.A.Orenstein and

M.Rueschemeyer (eds), Left Parties and Social Policy in Postcommunist Europe, Boulder, CO.: Westview Press,pp. 11–46.

Rotariu, T. and Popescu, L. (1999) “Poverty in Romania”, in Y.Atal (ed.), Poverty in Transition and Transition inPoverty: Recent Developments in Hungary, Bulgaria, Romania, Georgia, Russia, Mongolia, Oxford: Berghahn andParis: UNESCO, pp. 102–129.

Sabatier, P.A. (1998) “The advocacy coalition framework: revisions and relevance for Europe”, Journal of EuropeanPublic Policy, 51 (1):98–130.

DIMITRI A.SOTIROPOULOS 231

Sacchi, S., Bastagli, F. and Ferrera, M. (2002) “Fighting poverty and social exclusion in Southern Europe: the case ofItaly”, paper presented at the international seminar on “Fighting Poverty and Social Exclusion in Southern Europe:Dilemmas of Organization and Implementation”, Bocconi University, ISPI, Milan, Italy.

Scholz, O. and Tomann, H. (1999) “The role of social policy during the transformation in Central Europe”, in I.Collier,H.Roggerman, O.Scholz and H. Tomann (eds), Welfare States: East and West, Basingstoke: Macmillan,pp. 93–111.

Shalari, Z. (1999) “Albania”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Shopov, G. (1999) “Legislative reform of social assistance: the philosophy of changes” in G.Karasimeonov andE.Konstantinov (eds), Social Policy: Philosophy of Reforms, Sofia: Gorex Press and Fr. Ebert Foundation,pp. 69–90.

Skocpol, T. (1992) Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States,Cambridge, MA: Harvard University Press.

Sotiropoulos, D.A. (2002) “From an omnipresent and strong to a big and weak state: democratization and state reformin Southeastern Europe”, in D.A. Sotiropoulos and T.Veremis (eds), Is Southeastern Europe Doomed toInstability?, London: Frank Cass, pp. 63–74.

Sotiropoulos, D.A., Neamtu, I. and Stoyanova, M. (2003) “The trajectory of post-communist welfare statedevelopment: the cases of Bulgaria and Romania”, Social Policy and Administration, 37 (6):656–673.

Standing, G. (1996) “Social protection in Central and Eastern Europe: a tale of slipping anchors and torn safety nets”, inG.Esping-Andersen (ed.), Welfare States in Transition: National Adaptation in Global Economies, London: Sage,pp. 225–255.

Stoyanova, K. and Kirova, A. (1999) “Bulgaria”, in Consensus Programme, Sector: social assistance, national survey,final version, PHARE, European Commission, ADECRI/PARTEX (July).

Stropnik, N. (1999) “Slovenia”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Szalai, A. (1999) “Hungary”, in Consensus Programme, Sector: social assistance, national survey, final version,PHARE, European Commission, ADECRI/ PARTEX (July).

Tesliuc, E.D., Pop, L. and Tesliuc, C.M. (2001) Poverty and Social Protection System, Iasi: Polirom Press.Todorova, S. (1999) “Emerging poverty in Bulgaria”, in Y.Atal (ed.), Poverty in Transition and Transition in Poverty:

Recent Developments in Hungary, Bulgaria, Romania, Georgia, Russia, Mongolia, Oxford: Berghahn and Paris:UNESCO, pp. 77–101.

Wagener, H.-J. (2002) “The welfare state in transition economics and accession to the EU”, West European Potitics, 25(2):152–174.

Walby, S. (1999) “The new regulatory state: the social powers of the European Union”, British Journal of Sociology, 50(1):118–137.

World Bank (2000) Making Transition Work for Everyone: Poverty and Inequality in Europe and Central Asia,Washington, DC: The World Bank.

Zamfir, C. (ed.) (1999) Politici Sociale in Romania: 1990–1998, Bucuresti: Expert.Zamfir, C. (2000) “Politica sociala in Romania in tranzitie”, in E.Zamfir, I. Badescu and C.Zamfir (eds), Stare

Sociatatii Romanesti Dupa 10 Ani de Tranzitie, Bucuresti: Expert, pp. 13–33.

232 SOUTH-EASTERN EUROPE: POVERTY

Index

Note: page numbers in italics refer to figures or tablesage factors:

family benefits 170–1;GMI 237;poverty 100–1, 135n35, 155, 157;Rentas Mínimas de Inserción 182;unemployment 40;see also elderly poor

Agricultural Insurance Organisation (OGA) 13;benefit delivery 64;family benefits 53;pensions 37–8, 51, 52;social insurance 35

agricultural workers 207, 210, 212, 219, 224agriculture 9, 27Albania 272, 276alcoholism 275Amsterdam Treaty 3, 285Ancona municipality 88Armed Forces Movement 213Armenia 270, 271Arriba, A. 19Asia Minor Catastrophe 34associability 290

Baldini, M. 106, 107Balearic Islands 18, 151Balkans 276, 279–80Baltics 273, 279, 285Bank of Italy 106, 107Barr, N. 274, 275, 281Basic Law of Social Security, Spain 145Basque Country:

General Plans 150;home rule 189–90;minimum income schemes 176, 178, 179, 191–2;RMI 11, 18, 151

Bastagli, F. 14, 16Bauman, Z. 266Belgium 9Benassi, D. 105beneficiaries:

benefit administrators 10;characterization 245–6;commitments/obligations 184, 199–200n55;eligibility 64;family 177;minimum income schemes 191;Portugal 242–5;stigma 148;vulnerability 27;welfare funds 207, 210, 212

benefit administrators 10benefits:

contributory 12, 43;delivery 64–5;design 65–6;non-contributory 13;revaluation 67–8;rules/employers 63;targeted 22;see also means-testing;social security;welfare benefits

Berlusconi, Silvio 15, 24, 25, 84–5, 91, 110, 132n 12birth grant 54, 55black economy 9, 11;

see also informal economyBoeri,T. 101, 131n7Bomba, T. 19–20Budget Law, Italy 91Bulgaria:

extreme poverty 275;

233

income inequality 270, 271, 273;poverty alleviation 282–4;social assistance 280, 282–3

Caixas de Previdência das Casas do Povo 207Caixas de Seguros 212Caixas Sindicais de Previdência 207, 211Câmara Corporativa 210Capucha, L. 19–20Caramanlis, C. 70–1Carbonaro equivalence scale 95Care allowance 102care homes, children 282caring by women 205, 287Caritas 154–5, 175, 176, 187, 199 n 48Carta Magna 17, 147Casas do Pavo 207, 210, 211, 212Casas dos Pescadores 207, 210, 212Catalonia 11, 18, 179, 189–90Catania 88Catholic Church 85–6, 145–6, 187Centros Distritais de Segurança Social 215CERMI 175charity 75, 85–6child benefit 54–5childcare services 227, 259children:

in care homes 282;dropouts 254;labour markets 9;at risk 220, 254

Christian Democrats, Portugal 214, 234citizenship rights 178–9civil society organizations, Spain 175CLAs:

see Local Commissions for Social ActionCollective Agreements Act, Spain 145Commission for the Analysis of Macroeconomic

Compatibilities of Social Expenditure 88Commission of Inquiry on Social Exclusion 105Communist Party, Portugal 206, 213, 214, 234Community Cohesion Funds 215Community Development Methodology 212–13Community Support Fund 262n47, 262n49Comunidades Autónomas 147–9;

administration 176, 183;insertion activities 185–6;mesogovernment 192–3;minimum income schemes 19, 179, 182, 190–3;

poverty 150, 159;RMI 187, 189;TAPSIs 175

Conservatives, Greece 36coordination, open method 28criminality 273Croatia 267CSF (Parish Social Committees) 231Czech Republic 272–3, 275, 278

Deacon, B. 266, 284death rates 275decentralization:

Italy 15, 24, 84, 89;Portugal 232;Spain 23–4, 142, 153, 171, 193–4

Delors packages 286democracy:

Greece 12, 35–6;Portugal 20, 204, 213;Spain 141;welfare state 10–11

demographic factors 232–3,288Denmark 9developmentalism 145Diamantopoulou, A. 71disabled people 2, 44;

CERMI 175;Greece 35, 52–3;Italy 14, 105;Portugal 228, 259;poverty 274;social assistance schemes 36, 52–3;Spain 149, 150

dropouts 254drug abusers 220, 227, 259

Earned Income Tax Credit 79n30ECPF (Spanish Family Accounts Continuous Survey)

196–7n21Edis-Caritas 154–5education 123, 224–6, 254, 289–90EKAS (Pension Social Solidarity Supplement) 37, 51–2,

62, 65, 67, 71elderly poor:

Europe, Eastern 274;Home Help 58;housing benefit 56;Italy 14, 98, 100–1, 105, 133–4n24;

234 INDEX

marital status 135 n 38;pensions 2, 36, 100–1, 200 n 56, 233;Portugal 219, 227, 228;poverty 220;safety net 12–13;services 227–8

eligibility:beneficiaries 64;disabled people 44;GMI 237;RMI 113–14,115–16,122–3;social assistance schemes 2, 6, 12, 80 n 44, 87, 179–84emergency relief 34, 57

emigration, Spain 145employers, benefit rules 63employment process:

European Employment Strategy 250;full 4;poverty 223;social inclusion 224–6;women 8, 40, 205, 224, 233;see also labour force

Esping-Andersen, G. 266Estado Novo 19–20, 204, 207Estatuto do Trabalho Nacional 207Estonia 275, 279, 281ethnicity 75Europe, Central and Eastern 267;

EU 285, 291;income inequalities 273–4;poverty 271–2, 276, 290;poverty alleviation 278–82;social protection 266, 279

Europe, Eastern 4, 27–8, 274Europe, South-Eastern:

GMI 291;income inequality 269;informal economy 273, 287;poverty 272, 276;poverty alleviation 286–8;social allowances 267, 279–80

Europe, Southern:familialism 141;Fordism 5;informal economy 5, 10, 27;national health services 5;poverty 3, 6;safety net 6, 24–5;social indicators 7;

unemployment benefit 26;welfare developments 4–11, 141

European Commission 2–3, 23, 69–72, 194European Community 12, 23, 36European Community Household Panel data 42, 135 n 33,

153–4, 197n22, 220, 261n35;Greece 44, 45;households/poverty 156–7;Italy 99–100, 101

European Council 90European Employment Strategy Guidelines 250European Monetary Union 11European Programme to Fight Against Poverty 205, 213,

216, 234European Union:

Baltics 285;Europe, Central and Eastern 285, 291;National Action Plan for Inclusion 13, 23, 171;Portuguese presidency 251;social exclusion 2–3;social inclusion 11;social policy 1;welfare systems 285–6

EUROSTAT 44–7, 226, 228ex-convicts 220

FAINA (Support Fund for Insertion in New Activities)248

familialism 8, 9, 40, 141family:

beneficiaries 177;Greece 39, 40, 53–5;income 44, 260 n 5;minimum income schemes 182;social contacts 78n10;Spain 145;types 2, 219, 220;welfare 8;see also households

family benefits 28 n 6;age factors 170–1;contributory 55;fertility rates 109;gender equality 11;Greece 53–5;Italy 86, 102, 104, 105, 109;Portugal 207;Spain 160, 168

family size:

INDEX 235

Portugal 226–7, 232–3;poverty 46, 47, 49, 96, 97, 98, 100, 101–2, 223;social assistance schemes 15, 35, 36, 37, 53–5, 68–9;welfare benefits 152–3

fascism 86Federação Caixas de Previdência e Abono de Família

211Ferge, Z. 266, 284, 288Fernandes, R. 19–20Ferreira, L.V. 217, 261n31Ferrera, M. 127, 128, 291fertility rates 8, 36, 109Financing and Improvement Law, Spain 146FIPOSC (Fighting Poverty and Social Exclusion in

Southern Europe) 194fiscal police 123fishermen 207, 210Fordism 5, 12, 42Framework Law no. 328 (2000) 15, 89, 90, 91, 122, 128France 11Franco regime 16–17, 144, 145, 146fraud 247Fuero del Trabajo 144Fundo Nacional de Pensões 211

Galicia 189–90García Serrano, C. 155GDP 9, 20, 73, 92–4gender factors:

familialism 8, 9;family benefits 11;household heads 134n31;poverty 96, 134n31;unemployment 40, 225

General Directorate of Social Action, Childhood andFamily, Spain 177

General Education Law, Spain 146General Law of Social Security, Spain 146Greece:

benefit delivery 64–5;Conservatives 36;democracy 12, 35–6;disability benefits 35, 52–3;EKAS 37, 51–2, 62, 65, 67, 71;emergency relief 34, 57;ethnicity 75;European Community 12, 36;family 39, 40, 53–5;fertility rates 36;

history 77n3;Home Help programme 58;housing assistance 56;housing conditions 49–50;IKA 34–5, 37, 52;immigration 40–1, 49–50, 81n47;International Labour Organisation 34, 77n5;Labour Force Survey 40;Liberal government 34;military regime 35, 77n4;minimum income schemes 13, 69–75, 76, 80n40, 287;Ministry of Health and Welfare 53, 55;National Action Plan 38, 44, 57–60, 63–4, 70–1;national health services 36;National Social Care Organisation 38;New Democracy 70–1;OAE� 35, 50, 55, 59, 62–3;occupational schemes 11–12;OEK 56, 68–9;OGA 13, 35, 37–8, 51, 52, 53, 64;pensions 12–13;poverty 6, 33–4, 38–40, 44–9;poverty alleviation 34–8, 50–7, 59;public administration problems 24;redistribution of wealth 57;retirement benefit 51–2;safety net 3, 11, 75–6, 287;social assistance schemes 33–4, 43–4, 60–2;social exclusion 38, 74;social expenditure 78n13;social insurance 11–13;Socialists 37;trade unions 71, 78n12;unemployment benefit 37, 50, 58;welfare patronage 10;welfare state 11–12, 34–5, 43

Greek Orthodox Church 75group predation 9Guaranteed Minimum Income:

age factors 237;benefit calculation 237–9;Europe, South-Eastern 291;implementation 246–9;means-testing 238–9;National Minimum Income Commission 240–1;non-take-up 249;Portugal 205–6, 235–49, 252;Regional Social Security Centres 238–9;social integration contract 239–40;

236 INDEX

social policies 250;see also Reddito Minimo di Inserimento

gypsies 248, 274Gypsy Secretariat 175

Health and Social Welfare Services, Portugal 214health care 195n7, 207, 289–90HIV/AIDS victims 220, 227Hobsbawm, E J. 77n7Hoelscher, J. 289Home Help programme 58home working 9homeless people 220, 237Horizontes 2000 245, 247Household Budget Survey data 45, 47, 217households:

gender of heads 134ns31;income 177, 238;informal market 228;RMI 115, 118–19, 124–5;types 10, 27, 99–100, 226–7;see also family

housing assistance 13, 56housing conditions 49–50Hungary 275, 279, 281

IDS:see Institute for Social Development

IKA (Social Insurance Foundation) 34–5, 37, 52illegal earnings 273immigration 40–1, 49–50, 81n47, 151income:

family 44, 260n5;households 177, 238;non-monetary 248;poverty thresholds 68

income distribution 217–24, 221income inequalities:

cross-country comparisons 270, 271, 290;Europe, Central and Eastern 273–4;Europe, South-Eastern 269, 277

Income Support, UK 2Income Tax Act (Spain) 152Indicatore della situazione economica equivalente (ISEE)

88, 105–6, 113–14, 127, 136n47, n48Indicatore della situazione economica (ISE) 15, 106Industrial Employers Federation 20INE (Spanish National Institute of Statistics) 153informal economy 9, 11;

Europe, South 5, 10, 27;Europe, South-Eastern 273, 287;household services 228;OECD countries 79n32

INSERSO:see Institute for Social Services

insertion activities 120, 185–6, 243–4Institute for Social Development (IDS) 240–1, 263n56Institute for Social Security and Solidarity (ISSS) 249,

250Institute for Social Services (INSERSO) 147, 148, 149,

151, 195n10Instituto de Gestão Financeira da Segurança Social 215Instituto de Seguros Sociais Obrigatóios 206–7International Labour Organisation 34, 77n5International Monetary Fund 234International Standard of Poverty Line 94–5invalidity pensions:

civil 103, 105, 106–8;Greece 52;insurance-based 107;Italy 87, 103

Irs, Cles and Fondazione Zancan 113ISE:

see Indicatore della situazione economicaISEE:

see Indicatore della situazione economica equivalenteISSS (Institute for Social Security and Solidarity) 249,

250Istat survey 98, 109, 123Italy:

Budget Law 91;constitutional reform 90, 126;decentralization 15, 24, 84, 89;disabled people 14, 105;elderly poor 14, 98, 100–1, 105, 133–4n24;EMU 11;family benefits 86, 102, 104, 105, 109;fascism 86;Framework Law 15, 89, 90, 91, 122, 128;invalidity pensions 87, 103;maternity allowance 104–5;Mezzogiorno 24, 25, 128;minimum income schemes 14–15, 69, 87–8, 89, 103–4;municipalities 113, 114, 117–18, 127;National Action Plan 111, 132n11;national health services 86–7;National Social Fund 15;

INDEX 237

‘Pact for Italy’ 110;pension supplement 102–3;poverty 6, 94–102;poverty alleviation 84–7, 102–9, 129–30;regions 87–8;reinsertion programmes 119–21;safety net 3, 25, 84–5, 128;self-employment 106;social allowance 104;social assistance schemes 14–16, 84–5;social pension 103–4;Social Plan 89;taxation 108;trade unions 110;voting rights 86;welfare patronage 10;welfare state 85, 86–7, 94;Welfare White Paper 91;see also Reddito di Ultima Istanza;Reddito Minimo di Inserimento

Job Placement Offices 115–16

knowledge economy 233

labour force, women 8, 40, 205, 224, 233Labour Force Survey 40labour markets 5, 8–10;

education 224–6;flexible 285;Fordism 12;legislative reform 11;new entrants 44, 61–2

Latvia 281–2Lei de Bases da Segurança Social, Portugal 215, 216Liakos, A. 77n5, 77n6life expectancy 274–5Lisbon European Council 171Lithuania 279Local Commissions for Social Action (CLAs) 231, 241–

2, 262n47local policy, territorial pacts 137n63Lojas da Solidariedade 251

Macedonia 280Madrid 11, 18Mann, M. 138n74Manning, N. 286

Manpower Employment Organisation (OAE� ) 35, 50, 55,59, 62–3

marital status 135n38maternity allowance 15, 54, 104–5, 210Matos, G. 19–20Matsaganis, M. 11–12means-testing 10, 52–3, 127, 182, 183, 238–9, 290mesogovernments 176, 186–90, 187, 192–3Mezzogiorno 24, 25, 128Milan 88Milanovic, B. 281, 284minimum income schemes 1, 26–7, 191;

Basque Country 176, 178, 179, 191–2;beneficiaries 191;Comunidades Autónomas 19, 176, 179, 182, 190–3;family units 182;France 11;Greece 13, 69–75, 76, 80n40, 287;insertion activities 186;Italy 14–15, 69, 87–8, 89, 103–4;legislation 188;Portugal 20–2, 69, 287;poverty alleviation 72;regional 168, 171, 176–83, 180–1;Spain 18, 19, 145, 150–1, 168;see also Guaranteed Minimum Income;Reddito Minimo di Inserimento

minimum wage 170, 195n12, 213, 253Ministry for Social Affairs, Spain 17, 149, 186Ministry of Health and Welfare, Greece 53, 55Ministry of Labour and Social Affairs (MTAS), Spain

151, 154, 193Mitsotakis, C. 71mobility allowance 53modernization 4, 213, 220Moldova 270, 271, 272moonlighting 9Moreno, Luis 8, 19MSE (Social Employment Market), Portugal 231–2, 244–

5mutual aid associations 144, 207, 210

National Act of Social Services, Spain 149National Action Plan for Employment, Portugal 253,

262n50National Action Plan for Social Inclusion: EU 13, 23,

171;Greece 38, 44, 57–60, 63–4, 70–1;Italy 111, 132n11;

238 INDEX

Portugal 22, 251, 252;Spain 171, 172–6, 194

National Assistance, UK 2National Fund for Social Policies 116–17, 132n10national health services 5, 36, 86–7, 215National Institute for Social Insurance, Spain 16National Institute for Social Provision, Spain 144National Minimum Income Commission, Portugal 240–1,

242National Programme Against Poverty, Portugal 230–1,

234–5, 252National Programme to Fight Poverty (PNLCP), Portugal

217National Social Care Organisation, Greece 38National Social Fund, Italy 15nationality requirement 179, 182natural disasters 35Navarre 179neoliberalism 254, 284Netherlands 80n43New Democracy 70–1Nice Summit 3, 171–2Non-Contributory Pensions of the Social Security Law,

Spain 150non-governmental organizations 175, 176, 187, 198n40,

230–1

OAE� :see Manpower Employment Organisation

Obra Social 144Observatory on National Policies to Combat Social

Exclusion 2occupational programmes 11–12, 121OECD countries 1, 4–5, 79n32, 142OEK (Workers Housing Organisation) 56, 68–9Offe, Claus 290OGA:

see Agricultural Insurance Organisationold age services 228, 259;

see also elderly poorOmbudsman 79n31Onofri Commission 14, 88, 112, 119Organic Law on Social Security, Portugal 253

Parish Social Committees (CSF) 231part-time work benefit 58, 59Pascall, G. 286Patto territoriale 137n63pay-as-you-earn systems 269

peasants 248Pension Social Solidarity Supplement:

pension supplement 102–3, 159–60, 168, 170–1;see also EKAS

pensions 11;elderly 2, 36, 100–1, 200n56, 228, 233;Greece 12–13;low 44;minimum 37, 253;non-contributory 18, 151, 152, 160, 168;OGA 37–8, 51, 52;pay-as-you-earn 211–12;self-employment 77n6;social 2, 51, 103–4;supplements 51–2, 107–8;see also invalidity pensions

Perotti, R. 101, 131n7Plan Concertado, Spain 153Poland 275, 278, 281–2Popular Party, Portugal 247, 252–3Popular Party, Spain 151Portugal:

beneficiaries 242–5;caring by women 205;Christian Democrats 214, 234;Communist Party 206, 213, 214, 234;Community Cohesion Funds 215;coup 213, 232;decentralization 232;democracy 20, 204, 213;demographics 232–3;dictatorship 204, 210–11;disabled people 228, 259;economic growth 217;EEC 23, 204, 205, 206, 212, 228, 234;elderly pensioners 228;elderly poor 219, 227;employment 224–6;European Council 90;family benefits 207;family size 226–7, 232–3;GMI 205–6, 235–49;Health and Social Welfare Services 214;income distribution 217–24;industrialization 211;insertion programmes 243–4;Lei de Bases da Segurança Social 215, 216;Lojas da Solidariedade 251;minimum income schemes 20–2, 69, 287;

INDEX 239

minimum wage 213;modernization 213, 220;mutual aid associations 207, 210;National Action Plan for Employment 253, 262n50;National Action Plan for Inclusion 22, 251, 252;National Health System 215;National Minimum Income Commission 240–1, 242;National Programme Against Poverty 217, 230–1,234–5, 252;1933 Constitution 207;1976 Constitution 204, 213;Organic Law on Social Security 253;Popular Party 247, 252–3;poverty 6,19–20, 26–7, 204–5, 218–19, 233, 260n3;poverty alleviation 253, 255–6;poverty prevention 20, 251–2;presidency of EU 251;Regional Social Security Offices 241–2, 248;regions 219;safety net 3, 11, 232–6;self—employment 210, 212;single person households 226–7;social assistance schemes 19–22, 208–9;Social Council on Ministers 211;Social Democrats 234;Social Emergency Network 252;Social Employment Market 231–2, 244–5;social exclusion 235;social inclusion 3, 229–32;Social Insertion Income 22, 249;Social Network Programme 231, 250;social protection 207, 210, 211, 214, 215–16, 228,233–4, 257–8, 259;social security benefits 213–14;Social Security Law 229;social welfare institutions 20;Socialists 214, 229, 234, 235;trade unions 204–5, 212;unemployed 224–5;unemployment 235;welfare funds 207, 210;women in labour force 205

portuguese Family Budget Survey 261n35positive discrimination 229poverty 99, 136n50;

age factors 100–1, 135n35, 155, 157;at-risk rates 100;Comunidades Autónomas 150, 159;cross-country comparisons 267–8, 275;

disabled people 274;education 225–6;employment 223;Europe, Central and Eastern 271–2, 276, 290;Europe, South-Eastern 272, 276;Europe, Southern 3, 6;family income 260n5;family size 46, 47, 49, 96, 97, 98, 100, 101–2, 223;gender factors 96, 134n31;Greece 6, 33–4, 38–40, 44–9;income 68, 221;inequality 197n30;Italy 6, 94–102;Portugal 6, 19–20, 26–7, 204–5, 218–19, 233, 260n3;post-transition 270, 274–5;regions 95–6, 157–8;Russian Federation 272;social contacts 8;social exclusion 2–3, 11, 88;Spain 153–9;unemployed 101–2, 157;urban—rural divide 39–40, 46–7;vulnerable categories 220, 222, 223working poor 219, 224;see also elderly poor

poverty, types:absolute 47, 94, 98–9, 134n26, 271, 272;extreme 47–9, 275–6;persistent 223–4, 230, 275–6;relative 44–7, 94–5, 96, 97, 134n28

poverty alleviation 2, 42, 43;Bulgaria 282–4;Comunidades Autónomas 150, 159;Europe, Central and Eastern 278–82;Europe, South-Eastern 286–8;evaluation 281–2;Greece 34–8, 50–7, 59;Italy 84–7, 102–9, 129–30;means-tested 290;minimum income schemes 72;Portugal 253, 255–6;Russian Federation 281–2

poverty gap 99, 158, 197n31poverty prevention 4, 6, 20, 251–2poverty trap 65, 247power, state 138n74prison inmates 220Prodi, Romano 88, 112PROGRIDE programme 254

240 INDEX

protected special employment schemes 182–3public responsibility 250–1public utility programmes 121

Red Cross 175Reddito di Ultima Istanza 110–12, 126Reddito Minimo di Inserimento (RMI) 15–16, 24, 25–6,

109–13, 291;administration/implementation 114–17, 122, 123, 127–8;beneficiaries 117–19;eligibility 113–14, 115–16, 122–3;evaluation 110, 125–7, 288;exits 124–6;experimentation 85, 131n7;households 115, 118–19, 124–5;municipalities 113, 114, 117–18, 127;social inclusion 126;underground economy 124;see also Rentas Mínimas de Inserción

redistribution of wealth 57, 252Regional Social Security Centres, Portugal 238–9Regional Social Security Offices, Portugal 241–2, 248regions:

Italy 87–8;minimum income schemes 168, 171, 191;Portugal 95–6, 219;poverty 157–8;social services 147–9;Spain 18, 19, 142

rehabilitation programmes 123reinsertion programmes, Italy 119–21Rendimento Minimo Garantido 20–2rent subsidy 68–9Renta Activa de Inserción Laboral 160Rentas Mínimas de Inserción (RMI), Spain 150–1, 168,

176–8;age factors 182;Basque Country 11, 18, 151, 176–7;Comunidades Autónomas 187, 189, 192–3;regional 178–9

residence requirement 179, 182, 237retirement benefit 51–2, 170–1, 207;

see also pensionRevenu miminum d’insertion (RMI), France 11risk society 233RMI:

see Reddito Minimo di Inserimento;Rentas Mínimas de Inserción;

Revenu minimum d’insertionRoma 274Romania 280, 282–4Rome 88RSI (Social Insertion Income) 22, 249Ruiz-Huerta, J. 196n17Rural Areas and Social Development programme 254Russian Federation:

exceptionalism 273;household surveys 281;income inequality 270, 271;life expectancy 274–5;poverty 272;poverty alleviation 281–2;working poor 274

Sacchi, S. 14,16safety net 1, 3;

elderly poor 12–13;Europe, Southern 6, 24–5;Greece 3, 11, 75–6, 287;Italy 3, 25, 84–5, 128;perforated 287;Portugal 3, 11, 232–6;Spain 3, 11, 25, 142–3, 150, 161–7, 168, 169, 170–1,193–4;UK 2, 28n1

Scholz, O. 289schooling programmes 121, 123seamen 207seasonal work 9sectoral factors 9selectivity, generous 68–9self-employment 10;

Italy 106;pensions 77n6;Portugal 210, 212;social insurance 52, 59

Sicily 9Simitis, C. 37, 69–70single adults 98, 219, 220, 226–7Skocpol, T. 266Slovakia 275, 278–9Slovenia 272–3, 278, 279social allowances 104, 267, 279–80Social Assistance Fund 147social assistance schemes 1, 2, 131n1;

benchmarking 28;Bulgaria 280, 282–3;

INDEX 241

concentrated/dispersed/irrelevant 284;disabled people 36, 52–3;eligibility 2, 6,12, 80n44, 87, 179–84;family size 15, 35, 36, 37, 53–5, 68–9;Greece 33–4, 43–4, 60–2;Italy 14–16, 84–5;means-testing 10,14;Portugal 19–22, 208–9

Social Catholicism 146Social Charitable Protection Fund 144social citizenship 76, 151social contributions rebate 63Social Council on Ministers, Portugal 211Social Democrats, Portugal 234social dumping 142Social Emergency Network, Portugal 252Social Employment Market (MSE), Portugal 231–2, 244–

5social exclusion 2–3, 11, 38, 74, 88, 235social inclusion 3, 11, 126, 224–6, 229–32social inclusion process 13, 23;

see also National Action Plan for Social InclusionSocial Insertion Income, Portugal 22, 249social insurance 2, 4, 5;

compulsory 206;contributory 152;cross-country comparisons 280–1;Fordism 42;Franco regime 16–17;Greece 11–13, 35;non-contributory 152;Portugal 19–20;self-employment 52, 59

social insurance contributions rebate 58Social Insurance Foundation:

see IKASocial Integration of the Handicapped Law, Spain 149social integration programmes 120–1, 123, 239–40Social Network Programme, Portugal 231, 250social policy 1, 4–5, 250, 281social protection:

Europe, Central and Eastern 266, 279;expenditure 41, 214, 215–16, 228, 233–4;Portugal 207, 210, 211, 214, 215–16, 228, 233–4, 257–8, 259;redistribution of wealth 252

social security benefits 160, 168, 213–14, 234Social Security Law (Portugal) 229, 239social services 142, 147–9

social welfare 20, 275Socialists:

Greece 37;Portugal 214, 229, 234, 235

Sotiropoulos, D.A. 27, 28Soviet Union, former 270, 271, 276Spain:

Basic Law of Social Security 145;Carta Magna 147;Catholic Church 145–6, 187;civil society organizations 175;Collective Agreements Act 145;decentralization 23–4, 142, 153, 171, 193–4;democracy 141;disabled people 149, 150;economic development 141;ECPF 196–7n21;emigration 145;European Community 23;family 145;family benefits 160, 168;Financing and Improvement Law 146;Franco regime 16–17, 144, 145, 146;General Directorate of Social Action, Childhood andFamily 177;General Education Law 146;General Law of Social Security 146;General Plans 150;health services 195n7;historical background 143, 144–53;immigration 151;Income Tax Act 152;Institute for Social Services 147, 151;mesogovernments 176;minimum income schemes 18, 19, 145, 150–1, 168,181;Ministry for Social Affairs 17, 149, 186;Ministry of Labour and Social Affairs 151, 154, 193;mutual aid associations 144;NAPSI 171, 172–6, 194;National Act of Social Services 149;National Institute for Social Insurance 16;National Institute for Social Provision 144;NGOs 198n40;Non-Contributory Pensions of the Social Security Law150;pension supplement 159–60, 168, 170–1;Plan Concertado 153;Popular Party 151;

242 INDEX

poverty 6, 153–9;poverty gap 158;regional minimum income schemes 176–83;regional system of social services 147–9;regions 11, 18, 19, 142, 157–8;safety net 3, 11, 25, 142–3, 150, 161–7, 168, 169, 170–1, 193–4;Social Charitable Protection Fund 144;Social Integration of the Handicapped Law 149;Stabilization Plan 195n8;third sector associations 142–3, 146–7;trade union 195–6n13;unemployment benefit 17, 160;welfare state 16–19, 142;see also Basque Country;Comunidades Autónomas

Spanish Economic and Social Council 198n39Spanish National Action Plan for Employment 198n42Spanish National Institute of Statistics (INE) 153Standing, G. 266, 280–1state power 138n74stigma 148subcontracting 9subsidiarity 15, 53, 89, 194Subsidio de Desempleo Agrario 160subsistence benefits 279superwomen 194n1Sweden 9

tax credit refunds 58, 59, 63–4tax evasion 10, 11, 27, 287taxation, Italy 108Territorial Action Plans for Social Inclusion (TAPSIs) 175territorial pacts 137n63TFR (Trattamento di fine rapporto) 133n21third sector associations 142–3, 146–7Toledo Pact 17, 150Tomann, H. 289trade unions 10;

Greece 71, 78n12;Italy 110;mesogovernments 187;Portugal 204–5, 212;Spain 176, 187, 195–6n13;welfare funds 212

training programmes 121transition countries 27, 285–6Trattamento di fine rapporto (TFR) 133n21Turin 88

UK 2, 28n1, 79n30Ukraine 273, 274underground economy 5, 8, 9, 124unemployed, long-term 2, 44, 160, 253, 262n50unemployment:

gender factors 40, 225;Portugal 224–5, 235;post-transition 270;poverty 101–2, 157

unemployment benefits:Europe, Southern 26;Greece 13, 37, 50, 58, 59;Italy 11;Spain 17, 160

universalism 15, 68–9urban-rural divide 27, 39–40, 46–7, 286USA 79n30

voting rights 86

Wagener, H.-J. 288wages, low 58, 59, 63, 226, 274;

see also minimum wagewelfare benefits:

EU 285–6;family 8;family size 152–3;low-income 159–60, 168;minimum wage 170;targeted 22;transition countries 285–6

welfare developments 4–11, 141welfare funds 207, 210, 212welfare patronage 10welfare state 2;

Beveridgean 142;Bismarckian 5, 86, 142;democracy 10–11EU 285–6;Greece 11–12, 34–5, 43;Italy 85, 86–7, 94;post-communist 267;reforms 288;Socialist 268–9;Spain 16–19, 142

welfare tourism 179women:

double burden 205, 227, 287;fertility rates 8;

INDEX 243

labour force 8, 40, 205, 224, 233;poverty 219, 220;superwomen 194n1;working mothers 196n19

work, paid/unpaid 8, 40, 194n1, 227Workers Housing Organisation (OEK), Greece 56, 68–9working poor 219, 274Working Tax Credit 79n30World Bank 282, 288–90

Yugoslavia 274–5, 279–80

244 INDEX