Skip to main content

Poverty Reduction in Europe: Social policy and innovation

Final Report Summary - IMPROVE (Poverty Reduction in Europe: Social policy and innovation)

Executive Summary:
Poverty Reduction in Europe: Social Policy and Innovation (ImPRovE) is an international EU funded research project that brings together 12 outstanding research institutes and a broad network of researchers in a concerted effort to study poverty, social policy and social innovation in Europe. The ImPRovE project aims to improve the basis for evidence-based policy making in Europe, both in the short and in the long term. In the short term, this is done by carrying out research that is directly relevant for policymakers. At the same time however, ImPRovE invests in improving the long-term capacity for evidence-based policy making by upgrading the available research infrastructure, by combining both applied and fundamental research, and by optimising the information flow of research results to relevant policy makers and the civil society at large.
The ImPRovE project starts from three observations about the political and socio-economic context of the European welfare states. First of all, before the crisis, despite higher employment rates and economic growth, the EU social indicators show that nowhere any substantial progress has been made in combating relative financial poverty for the population at active age. Second, new (often small-scale) social policies and actions have emerged in the spatial and institutional margins of national welfare states. As part of its Flagship Initiative ‘European Platform Against Poverty’ in the EU 2020 strategy the Commission announced that it will work “to design and implement programmes to promote social innovation ...”. Third, the Lisbon strategy and the newly agreed EU 2020 poverty reduction target are embedded in the notion of social investment or active welfare state, assuming complementarity between growth, jobs and social inclusion.
In this context, the two central questions of the ImPRovE project are:
- How can social cohesion be achieved in Europe?
- How can social innovation complement, reinforce and modify macro-level policies and vice versa?
For answering these questions, the project evaluates the Lisbon decade and the impact of the economic crisis and austerity measures on poverty and inequality. In addition, we study the links between institutionalised macro level social policies and local initiatives of social innovation. By developing and evaluating policy scenarios that contribute to meeting the EU 2020 social inclusion target, we will also look at the future. Finally, we contribute to the development of new cross-national indicators in the areas of minimum income protection and poverty measurement.
More specifically, in this project we:
- Assess the validity, robustness, trends and interrelationships of the Europe 2020 social inclusion indicators
- Evaluate the effectiveness of tax-benefit systems;
- Design cross-country comparable reference budgets for six countries;
- Contrast targeted vs. universal (child) benefits and identify optimal policy designs of childcare provisions;
- Map the adequacy of minimum income protection schemes;
- Develop policy scenarios that could contribute to meeting the EU 2020 social inclusion target;
- Analyse the situation of non-standard workers, migrants, Roma and young people;
- Complement the quantitative analyses by 31 case-studies of social innovation;
- Examine how Europe 2020, the EU social agenda and cohesion policy can be brought closer together (at all levels).
For carrying out these tasks, the ImPRovE Consortium can rely on a broad network of internationally renowned researchers in the field of poverty, social policy, the labour market and social innovation. The project output is published in three books, and a high number of working papers and journal articles.

Project Context and Objectives:
After the post-war period, when welfare states were successful in reducing relative income poverty, the status quo has persisted for over forty years. This raises important questions about a) the measurement of poverty, b) the limits and effectiveness of employment and social policies, the role of c) the growing socially innovative local initiatives (with local here referring to different sub-national territories from regions, provinces, departments, municipalities down to neighbourhoods) and of d) the EU social agenda and cohesion policy. This is the intellectual framework of this proposal in which we establish the link between measurement, poverty trends and spatial distribution, the effectiveness of employment strategies, social protection and services and of the increasing role of social innovative local actions in order to draw policy relevant conclusions on how to reach the EU poverty reduction.

a) Assessing poverty trends and measurement issues
According to the approach that is dominant in the EU, poverty is relative, multidimensional and gradual (Atkinson et al, 2002: 78). Hence it is essentially an ambiguous concept that can be understood in many ways. This is also reflected in the Europe 2020 poverty reduction target. The Europe 2020 poverty reduction target, as agreed by the European Council of 17 June 2010, is defined broadly, in terms of people who are at risk of poverty and/or severely materially deprived and/or living in households with very low work intensity.
The statement in the project call that “the period of economic growth in the early 21st century has not led to a substantial reduction in poverty rates in Europe“ is, however, largely based on trends of relative financial poverty (at-risk-of-poverty) as measured in cross section by the 60 percent of median equivalised income. This raises some important further questions. Does the picture change when panel data, other poverty indicators, typical risk groups (children, youth, women, low skilled...) and special concentration are considered? What would have happened without the public policy reforms or indeed without the growth in employment? Could we have already expected positive consequences of new empowering practices, participatory innovations and social investment strategies?
More fundamentally, using the relative income threshold implies that at-risk-of-poverty is defined relative to the standard of living in each Member State separately. Hence, what is regarded as a minimal acceptable way of life depends on the prevailing conditions of life and the level of social and economic development in the country concerned. The European Commission framed its underlying rationale as follows: “the key challenge for Europe is to make the whole population share the benefits of high average prosperity” (EC, 2004). This is fundamentally different from the American poverty concept, where the official poverty line is based on the cost of a nutritionally adequate diet multiplied by a factor that takes account of non-food spending (up-rated in line with consumer prices) (Couch and Zaidi, 2010). European concern is indeed not confined to “severe deprivation of basic human needs, including food, safe drinking water, sanitation facilities, health, shelter, education and information”, as the United Nations rightly defines absolute poverty as a meaningful global concept (United Nations, 2009). Likewise, the European poverty notion is limited neither to situations where people suffer from a cumulation of disadvantage which cannot be reached by macro-policies (Dahrendorf, 1990) nor to instances of social exclusion in the sense of a permanent dependence on the State (Engbersen, 1991). Instead, it refers more broadly to people whose resources are so seriously below those commanded by the average individual or family in their country that they run the risk of being excluded from ordinary living patterns, customs and activities (Townsend, 1985).
Although we consider that the notion underlying the relative at-risk-of-poverty indicator goes to the heart of the European social model, it is evident that the relativity of poverty should not be overstated: one must take care to neither inflate the extent of poverty in rich countries, nor (and perhaps more importantly) deflate it in poorer Member States; the poverty notion must always relate to situations of severe economic hardship, including in the richer countries. This is especially so when the median income increases in times of rapid economic growth (and vice versa in times of severe recession).The enlargement of the EU has further underlined the need for a relative indicator of at-risk-of-poverty that is socially and politically acceptable. The at-risk-of-poverty threshold should be neither too high in rich Member States, nor too low in poorer ones. In order for it to retain its credibility, the indicator should accurately reflect what people minimally need, given prevailing local patterns of living, in order to be able to participate adequately in social and economic life. In the richer countries of the EU, a threshold set at 60 percent of median income seems to meet this requirement and hence it enjoys a rather broad legitimacy. However, research is urgently needed to assess whether this is also the case in the poorer Member States of Eastern and Southern Europe. Additionally, the current at-risk-of poverty thresholds neglect cross-national differences in the availability of public goods and services as well as income-in-kind (Tsakloglou et al., 2010). Similarly, the indicators do not taken full account of differences in needs across households and persons. These are important questions that should be investigated in depth, both from a scientifically and a policy perspective.
More generally, the strengths and weaknesses of the Europe 2020 poverty and social exclusion indicators should continuously be assessed both from a theoretical and an empirical perspective. Against the background of the emergence of new social policies and actions in the spatial and institutional margins of national welfare states it is especially important to develop transdisciplinary and comparative assessment methodologies. In other words, the key questions following from this can be summarised as follows:
1. Were poverty trends really disappointing and, if so, why?
2. How can existing poverty indicators be improved?
3. How can the adequacy of local social policies be assessed?

b) Employment, social protection and social inclusion: complementary goals or a trilemma?
The EU policy framework assumes complementarity between economic growth, employment and social inclusion. Countries that witnessed considerable employment growth are however not necessarily the ones that achieved corresponding reductions in poverty (OECD, 2008). Research tentatively points to three principal reasons : job growth has not sufficiently benefited the poor, in-work-poverty became a European phenomenon and the adequacy of social protection for those who remained outside the labour market declined ( Cantillon, 2010 ).
In the ‘Social Service Trilemma‘ Iversen and Wren (1998) argued that as a consequence of the transition from an economy dominated by manufacturing production to one dominated by services production, modern welfare states are no longer capable of being successful in enhancing both employment and equality within a tight budgetary framework: “... governments and nations confront a three-way choice, or ‘trilemma’, between budgetary restraint, income inequality and employment growth. Private employment growth can be accomplished only at the cost of wage inequality. Therefore, if wage equality is a priority, employment growth can be generated only through employment in the public sector – at the cost of higher taxes or borrowing ...” (Iversen and Wren, 1998: 508).
With more nuance, in an assessment of the Lisbon strategy Atkinson (2010) argued that complementarity between economic growth, employment and social inclusion can be achieved but not assumed: complementarity can only be assured with the proper design of policy. He contrasts two strategies to increase employment rates among less productive people: one is to lower the reservation wage, another is to reduce the cost of job creation. The first option increases the risk of poverty among working people, the second has a budgetary cost. Vandenbroucke expressed the question in the following forceful way: “the argument may be that social investment fundamentally and necessarily implied recommodification of individual citizen’s rapport with labour markets, with detrimental effect on the living conditions of vulnerable people... The truth is that policy makers... are confronted with a real dilemma here. How can one “make work pay” without increasing poverty, when resources are scarce?” (Vandenbroucke and Vleminckx, 2011).
The focus on activation, employment and on investment human capital could also have implied resource competition: ‘new’ investment policies (child care, education, work related spending, leave systems) could have crowded out more redistributive old social spending (unemployment benefits, health care...). Most countries have indeed seen significant changes in the structure of public social expenditures. Statistics show an increase of the relative importance of ‘new social spending’ on childcare and other work-related aspects compared to traditional transfer payments” (Vaalavuo, 2011). There are good reasons to assume that ‘new policies budgets’, more so than the traditional cash transfer, tend to flow to the higher income groups. They make welfare states more service-oriented, and services are after all considered to be less redistributive than cash transfers (Esping-Andersen and Myles, 2009). This is especially true when it comes to work-related expenses.
Arguably, investment states are thus confronted with two ambiguities (Cantillon, 2010).
▪ The first relates to activation and marginalisation of those who remain outside the labour market. Activation aims at social and economic inclusion of people who face difficulties in finding their way to the labour market. To do so, they are assisted by a broad range of policy instruments, ranging from in work benefits, reductions of taxes and job subsidies to individual counselling, working time flexibility and child care. On the other hand, ‘activation’ also implies the use of ‘sticks’ and more individual responsibility. Depending upon budget constraints and the design of the policy programmes this leads to a greater or lesser extent towards financial exclusion of those who are or cannot be activated. This raises questions regarding the design of policies that can mitigate potential exclusion of those who are not or who cannot be activated.
▪ The second ambiguity refers to the distributional impact of ‘new policies‘. If one wishes to support women’s labour force participation, then a policy must be employed to facilitate a combination of work and family life. Obviously, this policy will first benefit those who already participate in the labour process, in the hope that the others will follow. However, after more than 30 years of ‘women’s revolution’, activity rates of low skilled mothers remain significantly below that of their high skilled counterparts. As long as women’s position remains strongly socially stratified and as long as there is no adequate support for those for whom the combination of work and family life is (be it temporarily) not an option (for example, mothers and fathers with a disabled child, single mothers with difficult teenagers, sick or disabled parents) the new risk policy is bound to generate Matthew effects (Cantillon, 2010; Ghysels, 2010). In order to reach the 2020 poverty reduction targets and to implement successfully the Active Inclusion agenda the role of services in the distributional outcomes of European welfare states should be scrutinised and policy designs identified that are not biased against lower incomes.

In other words, key questions are:
1. Employment growth is an important anti-poverty strategy, at the heart of the EU 2020 strategy. What is the likelihood of success? How and under which conditions can employment help?
2. What is the role of social services, adequate income support and of local social policies?
3. How can complementarity between employment, social protection and social inclusion be achieved?

c) How can local social innovation complement, reinforce and modify institutionalised policies?
Interest in local forms of social innovation in disadvantaged areas has been building up since the crisis of the national welfare states in the late 1970s and its ongoing restructuring (Chambon et al., 1982; Nussbaumer and Moulaert, 2007; Klein & Harrison, 2007) and is reinvigorated in the continuing search for socially creative ways to overcome new and growing inequalities in Europe. This local focus is justified, firstly, by the spatial concentration of processes of social exclusion and those subject to them in disadvantaged areas and, secondly, by the observation that precisely in these circumstances socially creative experiments to overcome new and growing inequalities in Europe emerge.
Thirty years after its emergence and in the face of new and growing inequalities in Europe, it is important to take stock of the strengths and weaknesses of local forms of social innovation and the institutionalised forms of social investment and redistribution. Firstly, social innovation is especially well-suited to deal with territorial concentration of social problems in particular areas, cities or neighbourhoods. Furthermore, traditional instruments of social redistribution have difficulties in turning around the local concentration of poverty and social exclusion, because this typically requires a cross-sectoral, unified way of policy-making which can address the concatenation of different problems on both individual level (unemployment, low educational attainment, poor health) and neighbourhood or area level (regional economic decline, insufficient social housing, poor public services, crime and vandalism). Assessing successful innovation policies can therefore shed light on efficient policy mixes to turn ‘vicious cycles of local decline’ into a ‘virtuous cycle of local recovery’. This will typically require a combination of targeted area regeneration programs, local economic development strategies, investment in more locally based social services and social work with more traditional welfare state instruments (social housing policies, activation policies for at-risk-of-poverty young people, etc.).
Thirdly, social innovation can shed light on the limits and deficiencies of the social investment and activation policy discourse. Most policy initiatives to reduce social exclusion and poverty have been dominated by the view the paid work is the best route out of poverty. In this paradigm and policy setting, the most pressing social risks are likely to concern groups who have weaker access not only to the labour market but also to policies designed to increase employability. To address the problems of particularly vulnerable groups such as underground workers, migrants and asylum seekers, lone mothers or people with severe and enduring psychiatric problems, special measures are required. Because their needs are beyond generic programmes, these groups call for specially designed, social innovative and local policy measures (improved community care for people with socio-psychic problems, measures to avoid concentration of asylum-seekers, proactive labour inspection, etc.). Social innovation programs can be particularly helpful in gaining insight in how to lower access barriers for marginalised groups such as the homeless, juvenile delinquents, etc., which are hardly reached by the more traditional welfare schemes.
Social innovation measures, finally, mobilise the action potential or ‘agency’ of the poor by helping them to acquire control over the means to satisfy their own needs. Research into social innovation has revealed a degree of agency and creativity among the population in disadvantaged areas, which conventional social policies have not fully exploited (Moulaert, 2002). By tapping the local knowledge, skills and resources within difficult areas, local projects can help build the capacity of communities to acquire control over the means to satisfy their own needs. By providing locally needed services through the community (for example, family-support services, local health and community services, etc.) they can also enhance social control and restore a sense of pride and optimism. For these reasons, we propose to assess the effectiveness of local social innovative policies and actions. We will do so from the perspective on how they may complement, reinforce and modify macro level policies by addressing some of the latter’s aforementioned shortcomings.
In this context, the two central questions of the ImPRovE project are:
- How can social cohesion be achieved in Europe?
- How can social innovation complement, reinforce and modify macro-level policies and vice versa?

Project Results:
The results of the ImPRovE project are documented in about 60 Working papers, 31 Case studies and 3 edited volumes . As explained above, the ImPRovE project was driven by two questions: (1) how can social cohesion be achieved in Europe; (2) How can social innovation complement, reinforce and modify macro-level policies and vice versa? The project studied these questions in depth, while working also on underlying and closely related topics. This has resulted in a rich picture of trends in poverty and social cohesion in Europe, the adequacy and effectiveness of fiscal and social policies, and the evolution, challenges and interactions between macro-level social policies and local initiatives of social innovation. In addition, the project challenged existing indicators and paradigms, and has proposed several methodological innovations. Given the large research output, it is impossible to do justice to all research findings. In what follows we briefly summarise the main conclusions with regard to: (1) conceptual and methodological innovations; (2) trends in poverty and social exclusion; (3) policy implications.
Conceptual and methodological innovations
In order to better grasp the interrelationships between macro level social policies, social investment, and local initiatives of social innovation, we have developed a metaphor to think through the relationship between localized forms of socially innovative actions and initiatives on the one hand and the macro-level institutions of the welfare state and dialectical interplay between state institutions and civil society associations on the other (see especially ImPRovE WP 13/03 and WP16/11). The three paradigms can be represented as animals with specific characteristics that might allow us to better identify their potential relations and interaction. The social protection paradigm can be compared to a big and slow elephant, moving in a brittle changing context, but also able to live in a herd and cooperate. Social innovation can be represented as a butterfly, experimenting its new colours, egocentric beauty, lightness and agility. Its flight is frail and it lives for only one day. Finally, lions represent the social investment paradigm providing a specific re-interpretation of state intervention in a political-economic context increasingly dominated by the market logic and competition. Each ‘animal’ has its strengths, but can they co-exist with other animals on an equal footing? Each of the three paradigms can also be referred to central subjects: the national state can represent the social protection paradigm, as the main provider of welfare policy and measures. Civil society, social enterprises and local public institutions are the main promoter of participation and social innovation. The market is crucial in the social investment paradigm, in which public investments are oriented to support the labour market as the main mechanism for social inclusion. They can represent the paradigms’ dominant logics, but have not to be identified with them, as their role is central but not exclusive.
Local social innovation is not so much a new paradigm for social intervention that can potentially replace the welfare state, but a wave of initiatives that is part of an on-going spatio-institutional restructuring of welfare states, leading to a more complex rather than a completely different welfare mix. Rather than considering them as an alternative to other policy initiatives, they should be considered complementary, and interacting in complex ways with other policies. In addition, we developed a theoretical approach and methodological basis to study social innovation with a trans-disciplinary, mixed and comparative method; defined clear protocols to carry out the field work with indications to identify, analyse and compare case studies; and identified concrete research tools for the processing and interpretation of the results (cf. Deliverable 14.1). Still it is noteworthy that In terms of research traditions, it has been proven difficult in ImPRovE to let the ‘classical social policy research tradition’ interact with the ‘social innovation’ strand, mainly for conceptual reasons. Nonetheless, significant progress has been made in clarifying conceptual and methodological differences and stimulating interaction.
An important part of the ImPRovE project has focused on assessing the validity and methodological robustness of existing indicators of poverty and social exclusion, most notably the Europe 2020 poverty reduction indicator, while also developing new indicators, in particular comparable reference budgets. With regard to the first point, several papers have been written (including ImPRovE WP 13/01, 13/07, 13/08, 13/09, 15/01 and 15/16). First, the papers show that taking account of the time dimension in poverty research is crucial. For doing so, it is essential to have access to high-quality panel data, and to avoid breaks in series over time. The extension of the standard panel duration of EU-SILC from 4 to 6 (or 8) years would be a positive evolution from this perspective. Second, it is clear that the indicator of low work intensity can be strengthened in several respects, and Ward and Özdemir (2013) have offered several useful recommendations in this respect (ImPRovE WP 13/09). Finally, the move from the current indicator of material deprivation to the new one, would result in an improved identification of those confronted with the most severe forms of deprivation, even though it would result in a break in series. It is therefore advisable to keep on using the original indicator of severe material deprivation for monitoring progress to the EU 2020 poverty reduction target, in parallel with the new indicator.
Second, quite some resources have been devoted to developing and applying a method for identifying minimum required resources for adequate social participation in a comparable way across countries. An important outcome is a theoretical and methodological framework for developing cross-country comparable reference budgets in Europe (ImPRovE WP 13/02). In the project, we have used this framework to develop baskets of healthy food, suitable clothing, personal hygiene, accessible health care, rest and relaxation in six European countries (Deliverable D13.2) and explored the use of these reference budgets as a complementary poverty indicator. This shows that the at-risk-of-poverty threshold is not only much lower in the poorer EU countries in some abstract sense, but does also reflect different levels of income adequacy, as can be illustrated by reference budgets. Furthermore, families renting on the private market, families with children and young people are relatively worse off when poverty is measured with a reference budget indicator as compared to the at-risk-of-poverty indicator (ImPRovE WP 16/05).
When it comes to assissing poverty for vulnerable groups such as Roma, we found that improving the data collection system on the situation of the Roma at both the national and the EU level is a prerequisite for adequate programme planning and design. Currently very few comprehensive programmes to improve the situation of the Roma are adequately monitored by an external agent (Deliverable D10.1).
If social researchers aim at contributing to improving the reduction of poverty in Europe, their analyses should not only be focused on identifying problems and policy weaknesses, but should also assess the feasibility and poverty-reducing effect of alternative policy scenarios. By doing so, they may contribute to more effective policy programmes, based on the best available evidence. This is precisely one of the major purposes of the FP7 funded ImPRovE project. During the last year of the project, the focus has been on developing and applying a range of strategies for identifying more effective and equitable policy reform packages which reduce poverty, while taking financial work incentives and budgetary considerations into account. More in particular, we illustrated three strategies for identifying effective policy reform packages for reducing income poverty in Europe: (1) reversing regressive reforms and using measures of ‘poverty reduction / public budget trade-offs’ to identify efficient and effective policy reforms; (2) upscaling existing policies; (3) breaking down policies into their elementary parts and combining policy options systematically for identifying the most effective policy design. All three strategies make use of the European microsimulation EUROMOD. We illustrated them with case studies of the United Kingdom, Greece and Belgium, three countries with a very different welfare state trajectory, and social outcomes. We show that all three strategies are useful for identifying policy packages that work, and for producing evidence for more effective poverty-reducing policy reform packages (see especially ImPRovE WP 16/14). We are strongly convinced that the approach can have many policy applications, and has the potential to contribute, for instance, to the formulation of more effective Country-Specific Recommendations.
Assessing trends in poverty and social exclusion
When the ImPRovE project started, the first data on the impact of the crisis still had to be published. However, by making use of microsimulations, some trends could already be established. At the same time, an important goal of the project was also to better understand poverty trends in the past, and draw policy lessons for the future. Papers focused on studying poverty trends, mainly focusing on financial poverty, have found that:
• It appeared that in the good years prior to the crisis only the new Member States were successful in closing the virtuous circle: at-risk-of-poverty declined while employment rose and social spending dropped. However, some of the poorest countries have experienced the largest increase in poverty during the recent economic crisis. At the same time, vulnerable groups fared the worst, except for the elderly who saw their poverty risk decline over the past few years (e.g. Deliverable D1.1).
• The economic crisis and ensuing austerity policies have resulted in a serious decline in incomes in many of the countries hardest hit by the downturn. In these countries, the Great Recession also seems to have affected the composition of the population in poverty. The group at the bottom of the income distribution is younger than it was before the crisis, and is more likely to be on low-paid work or on unemployment. Furthermore, 2013 poor are found to be considerably poorer than those considered as poor in 2009 (e.g. ImPRovE WP 14/04)
• In addition, we found that while national poverty stagnated during 2005-2009, EU-wide poverty substantially decreased in the same period. In contrast, although poverty has increased between EU-SILC 2009 and EU-SILC 2011 when measured with national poverty lines, the crisis seems to have halted, but not (yet) substantially reversed the convergence trend of the lowest incomes in the EU towards the EU-wide median income (e.g. ImPRovE WP 14/05).
In addition, we invested in studying longitudinal dynamics at the micro level. This unveiled interesting dynamics, which can be summarised as follows:
• With regard to financial poverty, we showed that both during the 1990s and during the 2000s, longitudinal research shows that in both periods, income events and especially changes in the household head’s labour earnings seem to be highly associated with poverty transitions in all countries, but more in the Mediterranean countries, while demographic events seem to be relatively more important in Northern countries (ImPRovE WP 15/13).
• Broadening the picture to material deprivation, it is shown that the share and structure of those at high risk of chronic material deprivation and longitudinal income poverty differs considerable across Europe. In almost all countries under examination, lack of full employment by the individual or, especially, by the household’s reference persons leads to higher risks of chronic material deprivation and longitudinal poverty (ImPRovE WP 15/16).
• Turning to low work intensity, we found that there is a positive cross-national correlation between the level of low work intensity and the number of people living in a household with a persistently low work intensity, both as a share of the total population below the age of 65 and as a percentage of those living in a household with a low work intensity (ImPRovE WP 13/09).
• Finally, looking at all three indicators together, analyses of the panel-component of EU-SILC further showed that all three Europe 2020 poverty reduction indicators are affected by a considerable degree of genuine state dependence, according to which the past in itself influences the probability of someone experiencing the same problem again in the future. As a result, combating economic hardship today clearly reduces the problem tomorrow.
Finally, we assessed also the overlap between financial poverty and material deprivation from a cross-sectional point of view. Consistent poverty, i.e. being affected both by relative income poverty and material deprivation, is present in all European Union Member States. A measure of consistent poverty is useful for identifying the most vulnerable (D12.3 ImPRovE Working Paper 15/22). The paper showed that consistent poverty is present in all European Union Member States. There are important differences in what factors shape the comparison of living in consistent poverty compared to living in income poverty only or in material deprivation only. This finding suggests that those living in poverty or social exclusion are far from consisting a homogeneous group and so the most vulnerable can be better identified and monitored. Accordingly, B. Kis and Gábos propose that an alternative (overlap approach-based) measures of monitoring the effectiveness of social inclusion policies in the European Union to be introduced. This is put into practice in another ImPRovE Working Paper by Geranda Notten (ImPRovE WP 13/13). She finds that using consistent poverty as a benchmark, a richer and more nuanced picture emerges. Using only a single criterion such as income to assess cash transfer programmes will tend to under-state their success. In contrast, programmes emerge as much more successful when looking at whether they reach those who are both materially deprived and have low income, that is, those who are ‘consistently’ poor. In other words, consistent poverty emerges as a potentially more robust measure then just ‘material deprivation’ or ‘financial poverty’. At the same time, more methodological research on estimating the policy impact on material deprivation or consistent poverty would be most welcome.
It is clear that great advances have been made over the past couple of years in the area of poverty measurement. Nonetheless, important improvements in the measurement of poverty and social exclusion are still feasible, mostly by improving data quality. Panel duration is one factor, but being able to link EU-SILC information to other information at the micro level with regard to wealth, expenditures etc. is another example. Other work carried out in the context of ImPRovE points to the importance of improving information on differences in consumption patterns and the price of goods and services (Goedemé and Collado, 2016), and improving methodologies for identifying the minimum resources required for adequate social participation, in a comparative framework (Goedemé et al., 2015). Clearly, more can be done in these areas in the future.
Policy lessons
Most of the research efforts within the ImPRovE project were directed at better understanding how policies shape social outcomes, and what can be done to improve the effectiveness of policies to improve poverty reduction in the future. We have focused on many different types of policies. In what follows, we draw particular attention to: (1) the interaction between local social innovations and macro-level policies; (2) minimum income and targeted policies; (3) the separate and combined effect of labour market trends, fiscal, and social policies over the past few years; (4) and lessons with regard to designing more effective policy reduction strategies.
1. Local social innovation, macro level policies and the role of the EU
How are local social innovations combating poverty governed in Europe, and how do they interact with macro-level social policies? This is the central research question that motivates the ImPRovE research strand on social innovation. Social innovation is an increasingly important, but contested quasi-concept without a universally agreed definition. In the ImPRovE project, we have defined social innovation as actions and initiatives aimed at the satisfaction of social needs that are not adequately met by market and macro-level welfare policies (content dimension) through the transformation of social relations (process dimension), which involves empowerment and socio-political mobilization (political dimension linking the process and content dimension). They depend on the active involvement of civil society organizations, social entrepreneurs and/or local governments, which are seen as crucial sources of innovation. Although local social innovations work from the everyday life context from people living in poverty, they mostly mobilize resources, actors and instruments that are situated on supra-local scales and therefore have a ‘bottom-linked’ character.
We have explored and analysed the governance dynamics of local social innovations through 31 case studies in three policy fields: labour market activation, education for ethnic minorities and housing (with an emphasis on homelessness). All case study reports can be downloaded from the ImPRovE website. The case studies are located in eight different countries, namely Austria, Belgium, Brazil, Hungary, Italy, Spain, Sweden and UK. This includes the various types of welfare regimes (corporatist, liberal, social-democratic, transitional and familistic) and Brazil as an emerging welfare regime to contrast with the mature welfare regimes in Europe. Each case study report describes the socially innovative initiative and its genesis, assesses in what sense the initiatives are socially innovative, maps the institutional context, welfare mix and governance relations and discusses a range of governance challenges that were previously identified through a literature survey (e.g. mainstreaming, participation, fragmentation of welfare mix, diversity, etc.).
With regard to consolidation and diffusion of social innovation, the following conclusions and hypothesis can be deduced from the case studies (see in particular ImPRovE WP 16/11):
• The local and small dimension is considered the ideal dimension for social innovation to rise, even though we cannot exclude a priori its emergence at higher levels. The local dimension, closer to citizens, is considered as the best to meet their needs and specificities, but also to mobilise participation and resources and to conduct experimentations on the ground with minimum efforts and risk. But the local and small dimension can create ‘traps’ for social innovation. In fact, the processes of diffusion and consolidation need specific adequate conditions to be effective, like dedicated funds, time, politics’ attention, favourable policy arrangements, and governance assets.
• The consolidation and diffusion of social innovation is a stage in the innovation cycle at which most of our case studies – being relatively recent initiatives – did not get yet. However, we could identify several types of consolidation and distribution strategies: (1) public-sector driven; (2) driven by civil society organisations driven; (3) spontaneous processes of diffusion and consolidation; (4) private for-profit organisations driven processes of diffusion and consolidation. Each type can be associated with several opportunities and pitfalls for the upscaling, diffusion and consolidation of social innovation.
• Some obstacles to the consolidation and diffusion processes can be the following: a) vested interests and consolidated paradigms that do not consent to move political attention and resources to social innovative policies, new needs and weaker target groups; b) market competition and mechanisms of resources allocation that can be positive in the start-up phase, can prevent the consolidation of innovative initiatives before they prove their positive social potential; c) public retrenchment logics can prevent the investment on social innovation and threaten shortfalls in funding the consolidation and diffusion processes; d) weak welfare state institutions could be not able to assess the best initiatives and pick up the process of consolidation, provide enabling legal conditions and integrate the innovative initiatives within the whole welfare system; e) a weak third sector could be not able to communicate and share its own initiatives and positive results; f) ineffective governance processes and weak networks can prevent the circulation of ideas and information that contribute to spread social innovation.
In addition, we have shed more light on the role played by the EU in ‘social innovation’ (including shortcomings and possible risks) (e.g. ImPRovE WP15/21, 16/03 and 16/10). Most literature on social innovation focuses on the local level and does seldom take into account the European dimension. By adopting a diachronic and long-term perspective we have furthered our understanding of the emergence and developments of the EU social innovation agenda, the various understandings of the concept at the EU level, as well as instruments and resources made available.
In particular, we have linked the analysis on the EU level with the local level by exploring the usage of EU resources made in local socially innovative projects. We have shown, firstly, how the EU level is a key layer of multi-scalar social innovation systems, often able to directly sustain local projects, bypassing the national level. The EU supports social innovation both directly (by providing different kinds of resources for local socially innovative projects) and indirectly, by supporting European umbrella organisations operating in the field of poverty and social inclusion. Although EU support for place-based social innovation is significant, we have also highlighted it is not consistent throughout the whole life cycle of social innovation. From a societal point of view, our analysis has thus shed light on a number of shortcomings and risks of the EU’s interventions in relation to social innovation, which should be taken into account by both EU policy-makers and stakeholders in order to improve the effectiveness of available instruments.
Furthermore, ImPRovE research has shown the potential as well shortcomings with regard to institutionalizing and upscaling place-based social innovations; these are not only local level ‘butterflies’, with limited impact on nation-wide welfare state provisions, nor are they completely unrelated to the social investment paradigm. We have highlighted how the European Union (DG Employment) has attempted to create links between local innovations and higher scales of governance, for instance, by the development of the model of ‘social experimentation’, which intends to test local social innovations and social investment provisions with a view to mainstreaming and up-scaling the most successful ones into the broader welfare state.
2. Minimum income protection and targeted policies: trends and potential for EU involvement
Minimum income protection is generally considered an important tool for mitigating poverty by ensuring a safety net of last resort. Several discussion papers have been produced in which minimum income protection is presented as a policy area where increased EU social governance is both conceivable and needed. This holds particularly true for minimum income protection schemes aimed at jobless households. As Van Mechelen and Marchal show (ImPRovE WP 13/11), substantial progress has been made over the past decades in the inception of general social safety nets of last resort across the EU. However, convergence in social assistance benefit levels has been limited and in many countries assistance payments have not kept up with the development of average wages and median equivalent household income. As a result, levels of minimum income support have become increasingly inadequate in terms of their capacity to reduce poverty, if measured by 60 percent of median equivalent income. In their conclusion, Van Mechelen and Marchal suggest that the negative impact of external pressures associated with globalization and domestic challenges like ageing population, eroding traditional family structures, de-industrialisation and the shift of emphasis from social protection to activation and social investment, may have been stronger than the potential positive contribution of EU social policy. In addition, in a discussion paper that assesses activation strategies directed towards minimum income recipients in 20 EU Member states, Marchal and Van Mechelen demonstrate that behind the realities of these strategies seldom lies the notion of active inclusion as defined by the European Commission in its 2008 Recommendation (ImPRovE WP 14/07). Not only do social assistance benefit levels tend to fall short of the poverty line, many countries also adopt predominantly negative incentives to increase labour market participation rates, rather than enabling measures that promote human capital investment. The brunt of European countries rely heavily on financial and non-financial incentives such as strong ‘reasonable’ job definitions or severe sanctions for non-compliance to job availability requirements to encourage employment.
A bit more favourable picture emerges when one focuses on minimum income protection for workers. First, Marchal and Marx show that there is no evidence of a general decrease in minimum wages within the EU over the past decade (ImPRovE WP 15/18). Second, in countries where the relative minimum wage did decrease, this was often counteracted (or at least softened) through direct income support measures, like a substantial decline in income tax liabilities or increases in child support through the tax system. However, Marchal and Marx also point to the fact that in-work poverty has persisted, even in countries that have taken substantial efforts to support households on low earnings. The effectiveness of employment conditional earnings subsidies has also been questioned by Kenworthy (ImPRovE WP 15/10). Cross-country comparisons offers little support for a conclusion that the institutional configuration of the English-speaking rich nations with US- or UK-style employment conditional earnings subsidies is very effective at generating high and rising employment, high and rising incomes among low-end households, or low and decreasing relative poverty rates. Quite a few other affluent countries have done as well as or better than the UK and the US in recent decades.
In another paper, Cantillon, et al. make use of both survey data and simulated incomes of typical households. (ImPRovE WP 15/03) The paper points to three conclusions. First, even in countries with traditionally rather compressed wage distributions and strong social dialogue low gross wages are insufficient for families with children. Not unsurprisingly then, the disposable incomes of jobless households on social assistance fall short everywhere. Second, in the past two decades, in many of the countries considered here, the shortfall of minimum income protection packages for families with children has become increasingly bigger. In most cases this has not been occasioned by median incomes increasing faster than wages nor by deliberate cuts in benefit levels. Rather, there is evidence for declining gross low wages and, as a consequence, for increasing policy efforts required to lift minimum incomes of working and non-working families up to the poverty line. Third, in Europe, most of the rich welfare states started indeed to work harder: in most cases net low wages grew faster than gross low wages, and in some cases the rise in the gap between gross low wages and median incomes has been offset by increasing welfare state efforts. Yet, nowhere this was sufficient for closing the gap resulting from lagging gross low wages. For some countries we do also have indications of deliberate political decisions to reduce minimum income protection for jobless households compared to low wages. This was the case in Sweden in the whole period, Denmark in the 90’s, the UK and Austria in the 2000’s. This indicates severe and increasing structural difficulties to reduce poverty: as a result of downward pressures on low wages it seems no longer possible to successfully combine adequate minimum income packages for working and non-working families with children and reasonable incentives to work without substantial and increasing additional welfare state efforts. These conclusions are complemented and reinforced by a paper by Palme and Cronert (ImPRovE WP15/12) which documents in detail the Swedish policies.
Given the above trends, Cantillon et al. argue that if the EU2020 targets on the reduction of the EU population at risk of poverty or social exclusion are to be taken seriously, European minimum standards for social assistance and minimum wages seem to be a necessary step (ImPRovE WP 15/20). European minimum standards are the place to start when striving for a more performant EU level involvement in the field of social policy. Thereby, a broad approach should be taken, including principles for minimum standards in social assistance and minimum wages. Often it is not so perceived but, especially in the context of budget restraints, minimum wages and minimum income protection for jobless households are inextricably related to each other. To this end Cantillon et al. believe that time has come for a modest shift to “second order input governance”. More in particular, they propose to include policy indicators regarding minimum income protection sensu lato in the recently revised EU monitoring process of the European Semester.
3. The individual and combined effect of labour market trends, taxation, and social policies over the past decade
In order to better understand the policy options for the future, substantial attention has been paid to analysing past trends. In an in-depth study of recent trends in EU member states, Gábos et al confirm the negative relationship between individual employment and poverty (ImPRovE WP 15/06). More in particular, the results indicate that post-transfer poverty to employment elasticity has been around 25 percent on average in the EU in the period between 2004 and 2011. The elasticity of the employment-poverty ratio – while it looks sizeable for the whole sample on average – seems to have some variations across countries and periods. Furthermore, the analysis suggests that the crisis has resulted in very different employment trajectories across countries. Relatively larger changes in employment seem to have been accompanied with reverse trends in poverty rates: when employment increases, poverty declines in most of the cases (and vice versa, when employment declines, poverty will be on the rise). However, the authors found no clear evidence that the relationship between employment and poverty rates differed between the pre-crisis and the crisis periods.
Countries differ greatly in how a change in overall poverty is attributed to the combination of changes in the population share of those in jobless households and of changes in the poverty rates of both individuals living in jobless and non-jobless households.
- The declining share of persons in jobless households between 2005 and 2008 played a very important (relative) role in the large poverty drops observed in Poland, Slovakia, Lithuania, Ireland and Estonia. In very few of the countries had there been such a positive effect in the second period, though: in most countries the share of jobless households increased, contributing to an overall increase in poverty. Largest contributions of this factor to poverty change were measured in Estonia and Spain. In both of these countries the massive rise of poverty was being accompanied to a large extent by an increase of the share of jobless households.
- However, it is not only the structural effect that played a role in overall poverty change in the two periods. The increased poverty rate of the jobless was a significant contributor to poverty change in Germany between 2005 and 2008 and in Slovakia between 2008 and 2012. Moreover, changing relative poverty of the non-jobless households also matters a lot. In Poland and in Slovakia, for example, a considerable part of the decline in poverty was accounted for by decreasing poverty rates of the non-jobless households, while in Latvia, Finland, Greece, Sweden and Germany a sizeable part of the increase in poverty is attributable to increasing poverty rates of the non-jobless households. In the second period, countries like Spain, Greece, Estonia, Austria, Italy and Romania with poverty increases of more than 2 percentage points, the increased poverty rate of households having at least someone in work has played a role in this overall increase.
To summarize the authors conclude that the success of poverty reduction depends to a large extent on three factors: the dynamics of overall employment growth, the fair distribution of employment growth across households with different work intensity and properly designed social welfare systems to smooth out the income losses of those families who are, for some reasons, unable to generate sufficient income for themselves from the labour market. First results of analysing ‘post-crisis’ trends are not very encouraging in this regard. Using evidence from the EU-SILC, Justin van de Ven and associated expert Sarah Voitchovsky confirm that lower standards of living and higher risks of poverty and social exclusion are associated with lack of employment (ImPRovE WP 16/07). Therefore, the decline in unemployment rates since the trough of the recent Great Recession leads to lower levels of poverty and social exclusion and is very welcome. However, the corresponding employment recovery in Europe seems to have favoured primarily highly skilled workers to the relative detriment of those with lower qualifications, while there appears to be some deterioration regarding the "quality" of the new jobs created (higher rates of part-time employment, temporary contracts, etc). These trends, if continued in the later stages of the economic recovery, may hinder efforts at reducing inequality and (relative) poverty. Moreover, a number of differences are observed between core countries with persistent labour growth, core countries with disrupted labour growth and "new" member-states.
Even though poverty rends have been disappointing before the crisis, it does not automatically follow that social and fiscal policies are to blame. It may even be the case that policies had become more progressive, but negative labour market effects (cf. above) or other factors contributed to an increase, or lack of decrease in relative income poverty. To shed more light on this, Hills et al. examined the extent to which tax and benefit policy changes introduced in the period 2001-11 had a poverty-reducing effect (ImPRovE WP 14/03). In other words, the authors assessed whether the period was indeed a “missed opportunity” for policy changes to make a difference to poverty reduction since the Lisbon Treaty, given the general lack of improvement shown by poverty indicators. The analysis used the tax-benefit model EUROMOD and covered seven diverse EU countries: Belgium, Bulgaria, Estonia, Greece, Hungary, Italy and the United Kingdom. The results by countries show that compared to the 2001 system, the 2011 policy system was more effective in reducing the risk of poverty in Belgium, Estonia and the UK. However, policy changes clearly resulted in increased poverty against an anchored poverty threshold in Greece and against a floating threshold in Hungary. Overall, there was considerable variation between countries in how different policy instruments have been adjusted, and in the effects of these adjustments by income, by age and by household composition, showing the importance of understanding them together, rather than discussing just some in isolation.
In addition, the authors also separated the total effect of tax-benefit policy changes into i) structural changes as changes in the design of the tax-benefit system (e.g. introduction of a new benefit, change in the tax regime) and ii) indexation effects as changes to the policy parameters with monetary values (e.g. benefit amounts and tax thresholds), relative to general price and wage developments. They found that benefit amounts and tax thresholds were mostly increased by more than growth in prices in the period as a whole and, during the crisis in the second part of the period (2007-11), also stayed ahead of growth in average market incomes (as it lagged behind price increases). Hence, in this period the indexation effect was typically more effective in alleviating poverty and inequality than changes to the structure of policies. Thus, the analysis served to highlight the importance of indexation policies in general and called for greater attention to ensuring their adequacy over time, an issue that is sometimes insufficiently recognised by policy makers.
4. Designing more effective poverty reduction strategies
The above implies that effective employment policies may not be sufficient to lift substantial numbers of poor above the poverty line and the role of adequate social protection is likely to be indispensable. Under these circumstances, is it possible to provide adequate minimum incomes and minimum wages without distorting work incentives? Using the tax-benefit model Euromod for three countries (Belgium, Denmark and the United Kingdom), the results of ImPRovE Working Paper 16/08 by Collado et al. suggest that this may be possible, but only at a high fiscal cost. Hence, if policy-makers are indeed interested in reducing inequality and (relative) poverty, it seems essential that the relevant policies should focus on the drivers of rising wage and market income inequalities rather rely than simply or primarily on the corrective impact of social transfers.
Looking at the same problem from a different angle, one may wish to ask how cost effective may be various policies aimed to reduce poverty. In the framework of ImPRovE Working Paper 16/6 (forthcoming) Leventi et al. developed a new method to answer this question, and applied this to the case of seven EU countries with very diverse tax-benefit systems, geographic location and economic situation (Belgium, Bulgaria, Estonia, Greece, Hungary, Italy and the UK), using Euromod. The focus is on (common) proportional changes in a number of policy instruments commonly used directly or indirectly for the purposes of poverty alleviation: child benefits, minimum income components of social assistance, (lowest) income tax thresholds and minimum wages, as well as on a more general aspect of policy-making, namely the regular indexation of benefit levels and tax thresholds. The main conclusion derived from the policy simulations carried out is that there is no “one size fits all” policy option, since different instruments appear to perform better in different countries.
Goedemé et al. (ImPRovE WP 16/14) conclude that it is possible to significantly reduce poverty and to design more effective policy reform packages, while taking budgetary constraints into account. More in particular the authors show several ways in which research can contribute to identifying poverty-reducing poverty reforms, making use of the European microsimulation model EUROMOD. In an application to the United Kingdom, Sutherland and De Agostini show how reversing regressive reforms and using measures of ‘poverty reduction / public budget trade-offs’, as developd in the paper by Leventi et al. cited above, can be a useful strategy for identifying efficient and effective policy reform packages, taking budget constraints into account. In contrast, Leventi and Matsaganis show how the upscaling of an existing pilot scheme for a guaranteed minimum income in Greece, could lead to a substantial reduction in the poverty gap by raising the incomes of hundreds of thousands of very poor families, accounting for nearly 11% of population. Finally, Vandelannoote and Verbist show how a more effective in-work benefit scheme can be designed for Belgium, by breaking down policies into their elementary parts and combining policy options systematically to identify the most effective policy design, while taking account of work incentives (see also ImPRovE WP 16/15). The simulations show that more effective policy designs aiming to reduce poverty are available. However, it is evident that in order to have a substantial effect on poverty an increase in the resources devoted to poverty alleviation is essential.
We are convinced that these analyses can be put to use for producing better informed and more effective, evidence-based, policies. For instance, the strategies for designing more effective policy reform packages could prove useful for formulating Country-Specific Recommendations in the European Semester, given little can be hoped from one-size-fits-all general recommendations. At the same time, policy makers should not forget about the limitations of the setup of the approaches we propose: we focus exclusively on measures of financial poverty and short-term poverty effects; all simulations are subject to data and specification errors; and, maybe most importantly, the analyses are limited to policy options that can be simulated on the basis of income survey data and the microsimulation models currently available. In other words, some policy options remain out of sight not because they are irrelevant, but because of the toolbox we can use for the analysis. Nonetheless, we are strongly convinced that a wider application of the techniques we propose would mean an important step forward for the design of more effective policy reform packages in the future.

Potential Impact:
A. Main dissemination activities so far
As the ImPRovE website was set up at the beginning of the project, throughout the project it has become an important means of communication, for all ImPRovE partners, associated experts and advisory boards, as well as for the broader public. All improve results, working paperse and case study reports can be downloaded from the website ( Results are also communicated through LinkedIn and Facebook, and updates of the website are available through an RSS newsfeed. Apart from the website, ImPRovE makes use of a wide variety of tools and channels for disseminating the knowledge generated in the project to all relevant audiences. We briefly discuss the most important channels for dissemination.
1. Scientific publications
Early 2013 the ImPRovE Working paper series was launched. The series collects the most important output of the project in a format that is principally directed to the broader research community. In contrast to some of the (larger) deliverables, the ImPRovE working papers are more accessible as they deal with one topic at a time, while some deliverables are a collection of working papers. The working papers allow for a detailed description of the work performed and provide all the necessary background information for correctly understanding the scientific output of the ImPRovE project. As such, the working paper series does not only function as an important tool for dissemination, but also serves as the main memory of the scientific output of the project. In total, about 60 Working papers have been published. Please note also that several additional working papers have been published that were not initially programmed in the Description of Work (including 13/04, 13/10, 14/02, 14/06, 15/09, 15/19, 16/02 & 16/15).
The papers are also archived and distributed through RePEc, an online repository that ensures a wide dissemination and easy accessibility of the ImPRovE Working papers. Since February 2013, ImPRovE working papers have been downloaded more than 1,760 times through RePEc and over 29,600 times through the ImPRovE website. Google scholar reports about 161 citations to ImPRovE Working Papers (April 2016). Given all papers have been published rather recently, we believe this is a satisfactory result. The strongly increased number of downloads through the ImPRovE website, shows that the website and the working paper series are becoming well known among the research community and the wider public.
Table 1: ImPRovE Working Papers
Number Title (short) Author
1 WP13/01 The Evolution of Poverty in the European Union: Concepts, Measurement and Data Decancq, Goedemé, Van den Bosch & Vanhille
2 WP13/02 Towards a common framework for developing reference budgets in Europe Storms, Goedemé, Van den Bosch, Devuyst
3 WP13/03 The butterfly and the elephant: local social innovation, the welfare state and new poverty dynamics Oosterlynck, Kazepov, Novy, Cools, Barberis, Wukovitsch, Sarius & Leubolt
4 WP13/04 Money or Kindergarten? Distributive effects of cash versus in-kind family transfers for young children Förster and Verbist
5 WP13/05 Why has social security become less pro poor? Cantillon, Van Mechelen, Pintelon & Van den Heede
6 WP13/06 The effect of local labour market conditions on educational choices: a cross country comparison Tumino
7 WP13/07 Alternative vs. current measures of material deprivation at EU level Guio and Marlier
8 WP13/08 The indicators of inter-temporal material deprivation: a proposal and an application to EU countries D’Ambrosio
9 WP13/09 Measuring low work intensity – an analysis Ward and Ozdemir
10 WP13/10 Testing the statistical significance of microsimulation results Goedemé et al.
11 WP13/11 Trends and convergence of Europe’s minimum income schemes Van Mechelen and Marchal
12 WP13/12 Exploring the multi-level governance of welfare provision and social innovation Oosterlynck, Kazepov, Novy, Cools, Wukovitsch, Saruis, Barberis and Leubolt
13 WP13/13 Measuring performance: does the assessment depend on the poverty proxy? Notten
14 WP13/14 In-work poverty in times of crisis: do part-timers fare worse? Horemans and Marx
15 WP14/01 Components of income inequality and its change in EU countries, 2004-2010 Medgyesi
16 WP14/02 Can education bridge the gap? Education and the employment position of immigrants in Belgium Corluy and Verbist
17 WP14/03 A lost decade? Decomposing the effect of 2001-11 tax-benefit policy changes on the income distribution in EU countries John Hills, Alari Paulus, Holly Sutherland & Iva Tasseva (ISER)
18 WP14/04 Distributive Effects of the Crisis and Austerity in seven EU Countries Manos Matsaganis & Chrysa Leventi
19 WP14/05 Mountains on the move: recent trends in national and EU-wide income dynamics in old and new EU member states Goedemé, Collado, Meeusen
20 WP14/06 Estimating the poverty reduction effect of tax and benefit policies in Finland 1993-2013 using a microsimulation method Moisio, Lehtelä and Mukkila
21 WP14/07 A new kid in town? Social inclusion in European minimum income schemes Marchal and Van Mechelen
22 WP15/01 The interrelationships between the Europe 2020 social inclusion indicators Sara Ayllón &
András Gábos
23 WP15/02 Involuntary part-time employment: perspectives from two European labour markets Michail Veliziotis, Manos Matsaganis, Alexandros Karakitsios
24 WP15/03 The end of decent social protection for the poor? The dynamics of low wages, minimum income packages and median household incomes Bea Cantillon, Diego Collado, Natascha Van Mechelen
25 WP 15/04 The impact of child care costs and availability on mothers’ labor supply Daniela Del Boca
26 WP 15/05 The role of immigration policies for immigrants’ selection and economic success Irena Kogan
27 WP 15/06 Employment and poverty dynamics in the EU countries before, during and after the crisis András Gábos, Réka Branyiczki, Barabara Lange and Istvan Tóth
28 WP 15/07 Non-cash benefits from social housing in Europe: a comparative perspective Marcus Grabka and Gerlinde Verbist
29 WP 15/08 Tracking, schools’ entrance requirements and the educational performance of migrant students Jaap Dronkers and Roxanne Korthals
30 WP 15/09 The joint decision of labour supply and childcare in Italy under costs and availability constraints Francesco Figari and Edlira Narazani
31 WP 15/10 Do Employment-Conditional Earnings Subsidies Work? Lane Kenworthy
32 WP 15/11 The social legitimacy of differently targeted benefits Wim van Oorschot and Femke Roosma
33 WP 15/12 Trends in the Swedish Social Investment Welfare State:
‘The Enlightened Path’ or ‘The Third Way’ for ‘the Lions’? Joakim Palme and Axel Cronert
34 WP 15/13 Mobility into and out of poverty in Europe in the 1990s and the pre-crisis period Eirini Andriopoulou and Panos Tsakloglou
35 WP 15/14 A fresh look at an old question: is pro-poor targeting of cash transfers more effective than universal systems at reducing inequality and poverty? Abigail McKnight
36 WP 15/15
Local social innovation and welfare state restructuring: analysing their relationship Oosterlynck, Kazepov, Novy, Cools, Sarius and Wukovitsch
37 WP 15/16 Chronic material deprivation and long-term poverty in Europe in the pre-crisis period Fotis Papadopoulos and Panos Tsakloglou
38 WP 15/17
Welfare systems, governance and social innovation: case study country profiles of Austria, Belgium and Italy Oosterlynck, Kazepov, Novy, Cools, Sarius and Wukovitsch
39 WP 15/18 Stemming the tide. What have EU countries done to support low-wage workers in an era of downward wage pressure? Sarah Marchal and Ive Marx
40 WP 15/19 Towards social investment and social innovation in EU member states? Karin Heitzmann* and Florian Wukovitsch**
41 WP 15/20 Decent incomes for the poor: which role for Europe? Bea Cantillon, Sarah Marchal and Chris Luigjes
42 WP 15/21 The EU framework for social innovation Between entrepreneurship and policy experimentation Sebastiano Sabato, Bart Vanhercke and Gert Verschraegen
43 WP 15/22 Consistent poverty across the EU Anna B. Kis and András Gábos
44 WP 16/01 The effectiveness of income-tested benefits in good times and bad: lessons from EU countries Chrysa Leventi, Olga Rastrigina and Holly Sutherland
45 WP 16/02 Has the potential for compensating poverty by women’s employment growth been depleted? Rense Nieuwenhuis, Wim Van Lancker,
Diego Collado and Bea Cantillon
46 WP 16/03 The usage of EU resources in local social innovation Sebastiano Sabato and Gert Verschraegen
47 WP 16/04 The tsunamis of educational attainment and part-time employment, and the change of the labour force 1960–2010 Wiemer Salverda
48 WP 16/05 Can reference budgets be used as a poverty line? Tess Penne, Irene Cussó Parcerisas, Lauri Mäkinen, Bérénice Storms and Tim Goedemé
49 WP 16/06 Improving territorial cohesion: the role of stakeholders in OMC and cohesion policy Roberta Cucca and Yuri Kazepov
50 WP 16/07
Climbing out of Crisis: Characteristics of the newly employed in Europe Justin van de Ven and Sarah Voitchovsky
51 WP 16/08
The end of cheap talk about poverty reduction: The cost of closing the poverty gap while maintaining work incentives Diego Collado, Bea Cantillon, Karel Van den Bosch, Tim Goedemé and Dieter Vandelannoote
52 WP 16/09 The distributive effects of work-family life policies in European welfare states Hufkens and Gerlinde Verbist

53 WP 16/10 The integration of place-based social innovations into the EU social agenda Gert Verschraegen and Sebastiano Sabato
54 WP 16/11 Towards a more effective governance of socially innovative policies – first insights from the case studies Oosterlynck, Novy, Kazepov, Verschraegen, Saruis, Colombo, Cools, Cucca, Leubolt and Weinzierl
55 WP 16/12
The antipoverty performance of universal and means-tested benefits with costly take-up Alari Paulus
56 WP 16/13
The Europe 2020 social inclusion indicators: main conclusions of the ImPRovE project András Gábos and Tim Goedemé
57 WP 16/14 Scenarios for reducing poverty in Belgium, Greece and the United Kingdom Goedemé, Sutherland, De Agostini, Leventi, Matsaganis, Vandelannoote and Verbist
58 WP 16/15 The design of in-work benefits: how to boost employment and combat poverty in Belgium Dieter Vandelannoote and Gerlinde Verbist
59 WP 16/16 Improving poverty reduction in Europe: what works (best) where? Leventi, Sutherland and Valentinova Tasseva

Table 2: Access and download statistics of ImPRovE working papers through RePEc (March 2015-March 2016)
2015-03 144 160
2015-04 53 121
2015-05 70 103
2015-06 37 120
2015-07 52 114
2015-08 41 82
2015-09 76 99
2015-10 31 118
2015-11 51 141
2015-12 108 102
2016-01 28 70
2016-02 33 142
2016-03 111 108

During the second reporting period, we have also launched our series of ‘ImPRovE Case Studies’, in the context of work package 11, containing 31 case studies in total. So far, case studies have been downloaded less than the ImPRovE working papers, but most achieve also a considerable readership. All case studies can be downloaded from the ImPRovE website:
So far, publications in international peer-reviewed journals include:
Ayllón, S. and Gábos, A. (2016), ‘The Interrelationships between the Europe 2020 Poverty and Social Exclusion Indicators‘ in Social Indicators Research: 1-25. doi: 10.1007/s11205-015-1212-2
Goedemé, T. and Collado, D. (2016) ‘The EU Convergence Machine at Work. To the benefit of the EU’s poorest citizens?‘ in Journal of Common Market Studies:1-17. doi: 10.1111/jcms.12382.
Goedemé, T., Storms, B., Stockman, S., Penne, T., and Van den Bosch, K. (2015), ‘Towards Cross-country Comparable Reference Budgets in Europe: First Results of a Concerted Effort‘ in European Journal of Social Security (EJSS), 17(1): 3-31.
Goedemé, Tim; Van den Bosch, Karel; Lina Salanauskaite en Gerlinde Verbist (2013), ‘Testing the Statistical Significance of Microsimulation Results: A Plea’ in International Journal of Microsimulation, 6(3): 50-77
Manos Matsaganis & Chrysa Leventi (2014), The Distributional Impact of Austerity and the Recession in Southern Europe, South European Society and Politics, 19:3, 393-412, DOI: 10.1080/13608746.2014.947700
Marchal, S., Van Mechelen, N. (2015), A new kid in town? Active inclusion elements in European minimum income schemes. Social policy and administration – ISSN 0144-5596, p. 1-24.
Notten, G. (2015), ‘How Poverty Indicators Confound Poverty Reduction Evaluations: The Targeting Performance of Income Transfers in Europe‘ in Social Indicators Research: 1-18. doi: 10.1007/s11205-015-0996-4.
It is noteworthy we received the FISS/Intersentia Best Paper Prize 2014 for the article by Goedemé et al. (2015) on reference budgets. We also published an article in local journal journals (more details can be found on the ImPRovE website). In addition, several book chapters have been published, covering the three main languages of the EU:
Marchal, S., Van Mechelen, N. (2015), Nil novi sub sole? Actieve inclusie in Europese bijstabdsstelsels. Armoede en sociale uitsluiting: jaarboek 2015 / Dierckx, Danielle [edit.]; et al.-isbn 978-94-6292-253-2-Leuven, Acco. [published in the third reporting period]
Cantillon, Bea (2014), ‘Beyond Social Investment. Which Concepts and Values for Social Policy-making in Europe?’ in Cantillon, Bea & Vandenbroucke, Frank (eds.), Reconciling Work and Poverty Reduction. How Successful are European Welfare States?, Oxford: Oxford University Press: 286-318.
Decancq, Koen; Goedemé, Tim; Van den Bosch, Karel & Vanhille, Josefine (2014), ‘The Evolution of Poverty in the European Union: Concepts, Measurement, and Data’ in Cantillon, Bea & Vandenbroucke, Frank (eds.), Reconciling Work and Poverty Reduction. How Successful are European Welfare States?, Oxford: Oxford University Press: 60-93.
Oosterlynck, S. and Moulaert, F. (2014) ‘Innovation sociale en Flandre: le rôle catalysant des politiques d’innovation et urbaine’, p.175-192 in: Klein, J.L. Laville, J.L. and Moulaert, F. (eds.) L’Innovation Sociale. Les éditions ÉRÈS, Collection Sociologie économique.
Wukovitsch, Florian (2014): ‘Soziale Innovation zur Bekämpfung von Armut und Exklusion’. in Sedmak, C.; Gaisbauer, H.P.; Kapferer, E.; Schweiger, G.; Selke, S. (Eds.): Lesebuch Soziale Ausgrenzung. Wien: Mandelbaum Verlag, 119-122.

It can be expected that many more articles will follow after the finalisation of the project. In the Description of Work, we committed ourselves to publish at least three books. We are well on track for doing so.
A first book has been published by Oxford University Press (January 2014). In the book ‘Reconciling Work and Poverty Reduction in Europe, edited by Bea Cantillon and Frank Vandenbroucke, we examine the link between poverty on the one hand and labour market participation and the distributive capacity of welfare states on the other hand. The book focuses on the working-age population and the evolutions in Europe during the 'good economic years' before the financial crisis. This book project does not only build on research results of the ImPRovE project and ImPRovE Consortium. To an important extent it also draws on research that was undertaken by the coordinator in preparation of the ImPRovE research proposal and ImPRovE project.
The work for a second book is well under way. The book will contain the main conclusions of the ImPRovE project, in the format of an edited volume to which all partners will contribute. Talks with Oxford University Press are in an advanced stage and a contract will be signed soon. More information on the table of contents can be found in deliverable D18.8. We hope the book will be published before the end of the year.
Cantillon, Bea; Goedemé, Tim and Hills, John (eds.) (forthcoming) Improving poverty reduction in Europe. Lessons from the past, scenarios for the future, Oxford: Oxford University Press.
Previously, we have announced the publication of a book summarising the results of the case studies on local social innovation, carried out in WP11. The preparations for this book are also well progressing, as can be observed from deliverable D11.4 which contains a detailed overview of the book chapters. A book proposal has been submitted for consideration by the Policy Press. Publication is expected in the Winter 2016 / Spring 2017.
Oosterlynck, Stijn; Novy, Andreas and Kazepov, Yuri (forthcoming) ‘Local social innovation to combat poverty: a critical appraisal’.
2. Publications for wider audiences & Policy briefs
During the course of the project, some articles based on ImPRovE research have been published in the popular press (see complete dissemination list below). In January 2014, ImPRovE launched its newsletter series. Since then we had four edited issues (see Figure 1). The mailing has been enlarging from the beginning of the project, now it has 332 subscribed readers plus 1800 subscribers from the TÁRKI newsletter list. The mailing list includes research community members, experts from various stakeholders, members of the Commission and national institutions, ministries, MP’s.
The average open rate of the newsletter is about 30%, the click rate is around 10 % in average. Subscribers with most opens are from France, Belgium, USA, Netherlands and Greece.
The last online issue (2015/1) was published in October.
The ImPRovE project website, as well as the Facebook and LinkedIn social media profiles include all the necessary information for subscription and links to the last issues.
Figure 1: Newsletter 2015/1 published on in October 2015

In parallel with the newsletter, we launched two social media platforms (one on Facebook and the other on LinkedIn) to direct more visitors to the project website and to draw attention on the spot to the project topic, major events, deliverables with a regular, short and visible reminder for the research team, the widest audience, and the broader public. On LinkedIn we have 52 group members and we have launched several research news on new discussion and working papers. Our aim was to make a tool of collecting more subscribers to the project newsletter and attract more visitors to the project webpage. In accordance, the topics are short and include informative research findings. Our activity on Facebook is restricted to short news, to photo documentation of our events and to basic information on the progress of the project. We also publish here the actual issue of the newsletter. With regular news on working papers and policy papers, the page has more than 15 posts and 72 “likes” from the readers.
It’s clear that overall, the ImPRovE website has been our most effective communication tool, as is for instance witnessed by the high number of working paper downloads (over 29,500) through the website.
By now, three policy briefs have been published on the ImPRovE website:
• Policy Brief 01: Distributive effects of the crisis and austerity in seven EU countries: some key lessons
• Policy Brief 02: The effects of tax and benefit policy changes on poverty and income distribution, 2001 to 2011: Some lessons for the future
• Policy Brief 03: Decent incomes for the poor: which role for Europe?
The policy briefs have been downloaded via the ImPRovE website over 1,000 times in total (the first two over 450 times each, whereas the latter about 150 times). Initially more policy briefs were planned to be published. However, given the wide dissemination of the ImPRovE working papers and direct contacts between ImPRovE researchers and policy makers (most notably from the European Commission), we decided to invest more time in this direct communication, which we considered more effective for disseminating the main conclusions of the project.
3. Organisation of international conferences and seminars
The following international conferences and seminars were organised during the course of the ImPRovE project:
• First ImPRovE Conference: Improving poverty reduction in Europe, 14-15 November 2013, Brussels. (additional conference)
• ImPRovE Mid-term Conference: The distributive impact of policies before and during the crisis. Lessons for Europe 2020, 8 April 2014, Antwerp
• Session during international conference: Which social policy paradigm for the 21st century?, 9-10 October 2014, Berlin.
• International Seminar: Policy programmes to promote social inclusion among Roma, 28 November 2014, Budapest.
• Improving Poverty Reduction in Europe: Developing Scenarios for the future. Special workshop organised by ImPRovE and EMPOV, 20-21 January 2015, Antwerp.
• ImPRovE Final Conference: Improving poverty reduction in Europe: Lessons from the past, scenarios for the future, 3-5 February 2016.
During the final conference, three additional events were organised:
• “500 years of UTOPIA”, in collaboration with the Centre Pieter Gillis. This event was organised at the occasion of the anniversary of Thomas More’s famous book, and provided an excellent opportunity to link up with current thinking about utopia’s, and in particular a world without poverty. This allowed the ImPRovE Consortium and research to disseminate the project results to a wider audience than otherwise would have been the case.
• Roundtable and Stakeholder meetings in the context of social innovation. Over the course of one day and a half, one roundtable and two stakeholder workshops were organised on the final results of the ImPRovE social innovation research (WP4, WP11, WP14 and WP16). A wide range of people, both from academia, the public sector, NGOs and social entrepreneurs were represented and participated in the discussion. In addition, two other sessions were organised on the same topic, with high-standing international speakers.
• A seminar on the validity and methodological robustness of the Europe 2020 poverty reduction indicators. This seminar involved members of the Social Protection Committee’s Indicator Sub-Group, and most notably its chair, Rudi Van Dam. The seminar functioned as an occasion for discussing the main conclusions and recommendations of ImPRovE’s WP12, on the same topic.
Between 100 and 150 persons participated in each conference. For each conference participants included academia, NGOs, policy makers from various government levels (local, regional, national and European) and representatives from a wide number of countries, including the US, Canada, and Australia. The full programmes, presentations and some pictures are available from the ImPRovE website In other words, it is not exaggerated to say that the conferences have reached their goal of disseminating the most important conclusions of the ImPRovE project among key stakeholders and (EU) policy makers, as well as within the broader research community.
In addition, at several occasions we have presented the ImPRovE project and first results in meetings with staff members of the European Commission (especially DG EMPL). 13th June 2012 Stijn Oosterlynck presented the social innovation part of the ImPRovE project. In October 2012, the coordinating team (Bea Cantillon and Tim Goedemé) presented in a closed meeting, attended among others by Lieve Fransen, the entire project to DG EMPL. In April 2013 the coordinating team had another meeting in Antwerp with Commission staff member Istvan Vanyolos to discuss the development of references budgets in the entire European Union. Also, during our project meeting in Urbino Céline Thévenot acted as a discussant and presented recent results of research by DG EMPL in the area of poverty dynamics. Also in January 2015 a meeting was organised with representatives from DG EMPL of the European Commission, involving among others Lieve Fransen, Director Social Policies and Europe 2020. During this meeting we discussed the setup of Work Package 15 so as to inform the Commission of our plans and prepare an adequate dissemination strategy. Also, we were informed about information needs and hot topics on the agenda of the Commission so as to ensure an optimisation of the policy relevance of the work done in ImPRovE. In addition, there have been regular informal contacts between the coordinator and staff of DG EMPL.
Also after the end of the project, dissemination will continue. For instance, during the APPAM (Association for Public Policy Analysis and Management) International Conference 13-14 June 2016 in London, we have organised a session entitled ‘Improving Poverty Reduction in Europe: Identifying Policy Mixes That Work’, with presentations based on ImPRovE research.
4. Other national and international dissemination activities
Needless to say, ImPRovE results have also been disseminated through the participation in international and national conferences, debates and workshops. For a complete list of dissemination activities, see the table below. It is noteworthy to highlight that during the second reporting period ImPRovE was selected as a ‘best practice’ for a workshop on ‘Rethinking the role of social sciences (SSH) in H2020’, organised by the University of Bologna, who is not affiliated to this project. Also during the coming months ImPRovE results will be further disseminated through the publication of the ImPRovE books, journal articles and presentations at national and international conferences, including the FISS conference in June and the ESPAnet conference in September.
5. Conclusion
Overall, the dissemination strategy from the original work plan has been followed throughout the project, with the exception of the organisation of additional events and the production of less policy briefs than anticipated. We are convinced that the results of ImPRovE are disseminated effectively, given that research results are already used by various social actors, most notably the European Commission services. With over 29,600 downloads, the ImPRovE working papers seem to be widely picked up, with on average more than 500 downloads per working paper, even though a good number of papers have been published rather recently. Of course, further dissemination will continue after the project, given that the publication of international peer-reviewed journal articles requires some time before they are finalised and published. Also, the future publication of two books containing the main project results, with high-profile publishers, guarantees a continued and long-lasting dissemination of the principal conclusions of the project. It is also noteworthy that ImPRovE research will also result in at least one PhD thesis.
B. Potential impact and wider societal implications
As can be observed from the objectives of the ImPRovE project, and the summary of main results, the project has been focused not only on fundamental research, but also on generating tools and research results that are directly relevant for strengthening evidence-based policies. We hope to have made clear that with ImPRovE we did not only aim at high quality research, but also at actively contributing to 1) the policy-making process in the area of poverty, social policy and social innovation, especially at the European level; 2) the further development and integration of our field of study. We do so by organising regular contacts with the European Commission.
Some of the (modest, but noteworthy) impact we have probably had is visible in references to the ImPRovE project in the Commission’s texts on the Social Investment Package as well as a call to submit a project proposal on the development of a common methodology on reference budgets (VT/2013/041). UA-CSB took the lead of the European Reference Budgets Network pilot project, which was financed through the aforementioned call. From our point of view, the latter project can be considered a successful spin-off of the ImPRovE project. On the scientific side, the quality of the output of ImPRovE has been validated through several publications in international peer-reviewed journals. In addition, we were awarded the FISS/Intersentia Best Paper Prize 2014 for our work on reference budgets in this project, which also testifies to the quality of the output of ImPRovE. Given the high number of downloads of the ImPRovE Working papers (over 29,600 downloads by April 2016), the journal articles and book projects under preparation and planned dissemination activities in the near future, we are convinced that ImPRovE will have a lasting impact by informing future research on poverty, social policy and social innovation in Europe.
In addition, the ImPRovE Consortium actively liaised with other major projects in the field and stimulates cross-fertilisation of project results. For instance, a high number of other FP7 funded project were visibly present during the first ImPRovE Conference, and we contributed to the initiative of the COPE project to organise a joint conference on an Inclusive Europe in Berlin. In addition, the knowledge generated in ImPRovE has been used for starting up a project on reference budgets in Luxembourg and has actively contributed to a GRUNDTVIG project as well as the EMIN project . Several ImPRovE partners have now jointly submitted project proposals to the Horizon 2020 programme, building further on the knowledge gathered in the ImPRovE project.
Finally, we would like to highglight the strategies we have developed and illustrated for developing policy reform packages which are more effective in reducing poverty, while paying attention to budgetary constraints and financial work incentives. We are strongly convinced that a wider application of these strategies and tools could result in more solid, equitable and sustainable policy recommendations, and could help the formulation of evidence-based Country-Specific Recommendations. During the coming year, we will continue on this path, and actively liase with the European Commission to further disseminate the outcomes of the ImPRovE project.

List of Websites:

Bea Cantillon (Coordinator)
Phone: +32 3 265 53 98
Address: University of Antwerp - St. Jacobsstraat 2 (M. 177) - 2000 Antwerp - Belgium

Tim Goedemé (Co-coordinator)
Phone: +32 3 265 55 55
Mobile: +32 494 82 36 27
Address: University of Antwerp - St. Jacobsstraat 2 (M. 185) - 2000 Antwerp – Belgium

The contact details of all ImPRovE researchers can be found on the ImPRovE website.