My research interests can be divided into three broader themes: INEQUALITY, POVERTY and SOCIAL POLICY. Below you can find a short background of each of the themes together with abstracts of my journal articles, working papers, reports, briefs, and research in progress.
As part of my doctoral studies at the European University Institute (EUI) and as visiting scholar at the REAL Centre of the University of Cambridge, I analyse trends in intergenerational educational inequality and study mechanisms behind inequality in educational opportunities between children of differing socioeconomic backgrounds in sub-Saharan Africa (SSA).
In the first part of my thesis, I explore trends in association between parents’ and children’s educational attainment over the last three decades in SSA, with an aim to investigate whether educational expansion has been accompanied by a reduction of intergenerational inequality in primary school outcomes. Secondly, I study the role of national contextual factors in explaining cross-country-cohort variation in intergenerational educational inequality. This work is co-authored with Professor Fabrizio Bernardi
In the second part of my thesis, I study mechanisms behind inequality in educational opportunities between children of different socioeconomic groups in low-income settings. More specifically, I analyse whether unequal educational trajectories are driven by differences in children’s cognitive ability at school entry, by differences in family choices during the first years of primary school, or by differences in school quality. One of my case-studies is Ethiopia using data from the Young Lives international longitudinal study of childhood poverty
which follows the lives of 12,000 children over 15 years in four low- and middle-income countries.
In further analyses, I plan to study mechanisms behind the identified compensatory advantage by distinguishing between the effect of family behaviour and school effects. Incorporating the recently published school survey data in the Young Lives longitudinal data together with the RISE data will also allow me to analyse the interplay between the institutional context, policy change, and social stratification in children’s chances to attend and complete school in low-income contexts. This part of research I am carrying out in collaboration with researchers from the Research for Equitable Access and Learning (REAL) Centre
at the University of Cambridge
Research in progress:
Intergenerational educational inequality in sub-Saharan Africa for birth cohorts 1974-2003: Trends and macro-level explanations. With Fabrizio Bernardi.
Keywords: Trends in inequality; Educational opportunity; Primary school completion; Macro-level context; sub-Saharan Africa
While cross-country comparison on trends and determinants of inequality in educational opportunities is abundant in industrialized societies, evidence for low-income countries is scarce. In this article, we expand the geographic scope of this research and study intergenerational educational inequality – the association between parental socioeconomic status and children’s educational opportunities – in sub-Saharan Africa, a region characterized by economic development and educational expansion from 1990s onwards. First, we investigate trends in inequality in attendance and completion of basic education for birth cohorts 1974-2003. Second, we explore the role of macro-level characteristics in explaining country-cohort variation in inequality. We use data from 153 Demographic and Health Surveys and Multiple Indicator Cluster Surveys in 40 countries over 28 years (1990-2017) capturing over half a million children. Findings reveal that educational expansion in Africa did not equalize children’s chances to complete basic education. While inequality in attendance decreased, inequality in completion persisted. Cross-country variation of inequality in attendance is largely explained by macro-level differences in living conditions, public spending, private school enrolments, school fees, and colonial legacy. Variation of inequality in completion, by contrast, is better captured by teacher inputs and colonial history, underlining the importance of educational institutions in the stratification process.
Ability, family, or schools? Unequal educational trajectories in Ethiopia.
Keywords: Inequality. Cognitive ability. School transitions. Ethiopia.
This paper investigates the relationship between children’s early cognitive abilities and later parental decisions regarding investments in higher levels of education, focusing on disparities by parental socioeconomic status (SES). It tests whether differences in children’s initial cognitive abilities are reinforced or compensated by families’ educational investment decisions across three transition points of the educational cycle: upper primary, secondary, and higher education. The analysis is based on the Young Lives longitudinal study
which followed the lives of two birth cohorts born around 1994 and 2001 in four countries, including 3,000 children in Ethiopia. Findings point at three sources of inequality in educational opportunities. Firstly, poorer children on average develop lower cognitive abilities during childhood which negatively affects their later school transitions. This finding underlines the need for more policy efforts to equalize cognitive development opportunities early in life. Secondly, high-SES parents tend to compensate in case of early disadvantage at the primary school level. At higher levels of education, any remaining disadvantage is reinforced for all children, but SES gaps in children’s chances to make transitions increase at each level. Thirdly, low-SES children have meagre chances to transit to tertiary education, also when their initial endowments are high. This points at a loss of talent and a need to reduce existing barriers for children from low-SES families to transit to higher levels of education.
Education as an equalizer for human development? With Fabrizio Bernardi. Background paper for the UNDP Human Development Report 2019: Tackling Inequality in the 21st Century.
In this paper, we assess whether economic inequality is associated with negative health and social outcomes for countries of all levels of development, and whether higher levels of education at a societal level can moderate these negative effects. In the first part of the paper, we build on Wilkinson and Pickett’s ‘The Spirit Level’ and study macro-level associations between economic inequality and the following social problems: bad health outcomes, violence, early pregnancy, low educational proficiency among children, and social immobility. We expand the scope of the study by including countries of all levels of development. In addition to bivariate associations, we control for countries’ contextual factors to tackle the issue of possible spurious effects due to omitted variable bias. This approach can help identify the extent to which income inequality has a direct effect on social outcomes, and the extent to which this effect is absorbed by other underlying factors such as the level of economic development, government spending on welfare, and countries’ political and cultural context. Once the association between income inequality and social outcomes is established, we estimate the extent to which education moderates these associations.
In the second part of this paper, we look at the relationship between inequality, educational opportunities, and social outcomes at an individual level. We present results from the literature on social mobility and discuss the extent to which schooling and education can be regarded as an equalizer of intergenerational mobility chances in countries of high human development. The available evidence suggests that schooling is indeed an equalizer, although upper class families consistently manage to avoid downward social mobility for their children. Education is then the elevator that moves up for children of low socio-economic status (SES) families if they succeed in school. It does not, however, move down in case of failure for children from high SES families.
UNDP Human Development Report (HDR) Consultation Meeting, 23 April 2019, Qatar Foundation/WISE, Doha
Advisory Board Member for the 15th UNICEF Innocenti Report Card series, 2018: An Unfair Start: Inequality in Children’s Education in Rich Countries. Innocenti Report Card No. 15, UNICEF Office of Research – Innocenti, Florence.
Most of my work on poverty has been carried out at the UNICEF Office of Research where I worked between 2011 and 2014 as a social and economic policy consultant under the supervision of Prof. Chris de Neubourg and Prof. Sudhanshu Handa.
Whilst at UNICEF, together with a team of researchers I developed a methodology for measuring and analysing child poverty called Multidimensional Overlapping Deprivation Analysis (MODA)
. This methodology is currently used by many UNICEF offices and their counterparts to measure child poverty in all its dimensions, and to facilitate monitoring of trends towards reaching target 1.2 of the Sustainable Development Goals (SDGs). Upon requests from UNICEF regional and country offices, I was also responsible for leading workshops on the measurement of multidimensional child poverty. During regional workshops organized by UNICEF, I gave presentations on the measurement of poverty to UNICEF staff members and government counterparts in English and French. My tasks also included providing technical assistance to UNICEF’s country offices of Ethiopia, Tunisia, Senegal, and Madagascar where I assisted in the development of country-specific methodologies to measure child poverty, data analysis, and writing of reports. This work involved cooperating with the respective UNICEF country office staff and their counterparts from ministries, statistical offices and national research centres to ensure the applicability of the methodology to the local contexts. I continued research on the measurement and analysis of poverty at the Economic Policy Research Institute
(EPRI) where I worked between 2015 and 2016 under the supervision of Prof. Chris de Neubourg
. In my role as a social policy analyst at EPRI, I gave technical assistance and training on the measurement of poverty and multidimensional deprivation in Algeria, carried out a child poverty analysis in Kenya using a mixed-methods approach which included a qualitative research component, and was involved in other research projects related to poverty and social policy
Analysing Multidimensional Child Poverty in Sub-Saharan Africa: Findings Using an International Comparative Approach. With Marlous de Milliano. Child Indicators Research, 11(3), 2018, pp. 805-833.
Keywords: Child poverty. Multidimensional deprivation. Child rights. Sub-Saharan Africa.
This study provides with a first indication on the number of multidimensionally poor children in sub-Saharan Africa. It presents a methodology measuring multidimensional child deprivation within and across countries, and it is in line with the Sustainable Development Goal 1 focusing on multidimensional poverty by age and gender. Using the Multiple Overlapping Deprivation Analysis (MODA) methodology, the study finds that 67% or 247 million children are multidimensionally poor in the thirty sub-Saharan African countries included in the analysis. Multidimensional poverty is defined as missing two to five aspects of basic child well-being captured by dimensions anchored in the Convention on the Rights of the Child, namely nutrition, health, education, information, water, sanitation, and housing. The analysis also predicts the multidimensional child poverty rates for the whole sub-Saharan African region estimating 64% or 291 million children to be multidimensionally poor. In comparison, monetary poverty rates measured as less than USD 1.25 PPP per capita spending a day and weighted by the child population size finds 48% poor children. The results of this study highlight the extent of multidimensional poverty among children in sub-Saharan Africa and the need for children to have a specific poverty measure in their own right.
Child Poverty in the European Union: The Multiple Overlapping Deprivation Analysis Approach (EU-MODA). With Y. Chzhen, C. de Neubourg, and M. de Milliano. Child Indicators Research, 9(2), 2016, pp. 335-356.
Keywords: Child poverty. Child well-being. Multidimensional poverty. Poverty and deprivation overlap.
Poverty has serious consequences for children’s well-being as well as for their achievements in adult life. The Multiple Overlapping Deprivation Analysis for the European Union (EU-MODA) compares the living conditions of children across the EU member states. Rooted in the established multidimensional poverty measurement tradition, EU-MODA contributes to it by using the international framework of child rights to inform the construction of indicators and dimensions essential to children’s material well-being, taking into account the needs of children at various stages of their life cycle. The study adds to the literature on monetary child poverty and material deprivation in the EU by analysing several age-specific and rights-based dimensions of child deprivation individually and simultaneously, constructing multidimensional deprivation indices, and studying the overlaps between monetary poverty and multidimensional deprivation. The paper demonstrates the application of the EU-MODA methodology to three diverse countries: Finland, Romania and the United Kingdom. The analysis uses data from the ad hoc material deprivation module of the EU-SILC 2009 because it provides comparable micro-data for EU member states and contains child-specific deprivation indicators.
Lost (in) Dimensions: Consolidating progress in multidimensional poverty research. With C. de Neubourg and M. de Milliano. Working Paper No. 2014-04, UNICEF Office of Research, Florence.
Keywords: Child poverty. Child well-being. poverty. poverty reduction.
Identifying, locating and profiling the poor and deprived individuals in a society are the most basic imperatives for good social policy design. Understanding why people are, and remain, poor is the next analytical step. Multidimensional poverty and deprivation estimates are important new tools in this undertaking. This paper reviews the insights of various contributions from research into multidimensional poverty and deprivation and combines them into an internally consistent framework. The framework adds an important element by emphasising that people may experience various types and forms of poverty and deprivation simultaneously. The experience of poverty is often multifaceted and deprivations are interrelated in many cases. This highlights the necessity to clearly separate the different concepts of poverty and to study their overlap.
Keywords: Child well-being. Comparative analysis. Poverty.
Multiple Overlapping Deprivation Analysis (MODA) is a methodology developed by UNICEF which provides a comprehensive approach to the multidimensional aspects of child poverty and deprivation. MODA builds on earlier multidimensional poverty studies and encompasses a large set of tools ranging from deprivation headcounts in single dimensions via multiple overlap analysis to multidimensional deprivation ratios and their decomposition. The MODA methodology places the child at the heart of the analysis and concentrates on those aspects of well-being that are relevant for the children at particular stages of their lives. Moreover, the analysis indicates which deprivations children experience simultaneously.
Step-by-Step Guidelines to the Multiple Overlapping Deprivation Analysis (MODA). With C. de Neubourg, J. Chai, M. de Milliano, and Z. Wei. Working Paper 2012-06, UNICEF Office of Research, Florence.
Keywords: Child well-being. Comparative analysis. Poverty. Statistical methodology
Multiple Overlapping Deprivation Analysis (MODA) is a UNICEF methodology which provides a comprehensive approach to the multidimensional aspects of child poverty and deprivation. MODA builds on earlier multidimensional poverty studies and encompasses a large set of tools ranging from deprivation headcounts in single dimensions via multiple overlap analysis to multidimensional deprivation ratios and their decomposition.
Child Poverty in Kenya: A Multidimensional Approach. With E. Elezaj and C. de Neubourg. 2017, Kenya National Bureau of Statistics (KNBS) and United Nations Children’s Fund (UNICEF).
This study provides estimates of child poverty in Kenya in 2014 using a multidimensional approach, fulfilment of children’s basic needs and rights. It identifies the most vulnerable children, where the most deprived children live, the relationship between different types of deprivation that children experience, factors associated with child deprivation and poverty, and compares children’s multidimensional deprivation with monetary poverty. The report also compares the situation of children in realizing their rights in 2014 with 2008- 09 to shed light into the progress achieved in child poverty reduction. The methodology used in the study allows the generation of evidence to track progress in the Sustainable Development Goal (SDG) 1.2 on poverty reduction in all its dimensions for children, and SDG 10 on inequality reduction. Child poverty in this study is defined as deprivation in three to six dimensions. The indicators and dimensions used to measure child poverty were selected through a participatory and extensive consultation process involving the Kenya Bureau of Statistics (KNBS), sectoral experts from ministries in Kenya, sectoral specialists from UNICEF, and the Economic Policy Research Institute (EPRI). The study included a qualitative research component to identify the barriers in provision and demand of basic services and hurdles that children and their parents face in accessing them. The fieldwork was conducted during August 2016 in three selected counties, Turkana, Kakamega, and Kitui, on the dimensions of nutrition, health, education, water, and sanitation. Key informant interviews and focus group discussions were held with duty bearers and claimholders represented by parents, health facility personnel, community health workers, teachers, and county government representatives.
Multidimensional Child Deprivation Trend Analysis in Ethiopia. Further analysis of the 2000, 2005 and 2011 Demographic and Health Surveys data. With Kibur, M., Bitew, M., Gebreselassie, T., Matsuda, Y., Pearson, R. 2013, Calverton, MD: ICF International and UNICEF
This child-focused deprivation analysis sheds light on child poverty in Ethiopia, measuring child deprivation by using a number of dimensions of survival and development. It presents how different dimensions other than income poverty affect child well-being by using indicators from the Ethiopian Demographic and Health Surveys (2000, 2005 and 2011) and matching those to the rights contained in the CRC. The study measures the levels of child deprivation for the under-five child population and assesses overall progress in child deprivation reduction in Ethiopia over the years 2000 to 2011. The results show that while the deprivation incidence has decreased significantly in almost all dimensions between 2000 and 2011, the joint distribution of deprivations reveals that the percentage of children experiencing several deprivations at a time has decreased only marginally. In 2011, almost all children (94 per cent) still suffered from at least two deprivations considered as a threat to their survival or development. The average deprivation intensity was very high, children on average experiencing 3.8 deprivations at a time. This, however, is slightly lower compared to 2000 when children were on average deprived in 4.5 out of all six dimensions analysed. The deprivation overlap analysis shows differences in the extent to which the analysed sectors overlap, and reveals that children do not suffer from the same combinations of deprivations across regions. The study reveals significant disparities in multidimensional child deprivation levels between rural and urban areas and among regions. The highest child deprivation rates in 2011 were in Afar, SNNPR, Oromiya and Somali regions, while the lowest were in Addis Ababa, Harari and Dire Dawa. The highest decrease in the child deprivation level since 2000 has occurred in Amhara, Tigray, and Beneshangul Gumuz. Overall, the study aims to encourage the integration of child-specific needs into national poverty reduction strategies.
Cross-country Comparison of Multidimensional Child Deprivation Incidence and Intensity in sub-Saharan Africa. With Marlous de Milliano. MODA In Brief 4. Office of Research – Innocenti.
Multidimensional Child Deprivation and Monetary Poverty in sub-Saharan Africa. With Marlous de Milliano. MODA In Brief 7. Innocenti Working Paper No. 2014-19.
MODA database (last inserted data set: December 2014) With M. de Milliano, Y. Chzhen and C. de Neubourg. UNICEF Office of Research.
This web portal was designed at the UNICEF Office of Research between 2012 and 2014. It contains the following outputs and resources:
Cross-country Multiple Overlapping Deprivation Analysis for Children (CC-MODA) analyzing child deprivation for low- and lower-middle income countries according to internationally accepted standards of child well-being, utilizing internationally comparable datasets that contain child-specific information, i.e. DHS and MICS. The 'CC-MODA Technical Note' describes the technicalities of this analysis.
National MODA analysis for Children (N-MODA), an application of the MODA methodology to specific national contexts with customized dimensions, thresholds and indicators, utilizing richer information available from national datasets. This section is based on the input of countries carrying out N-MODA.
MODA for the European Union (EU-MODA) using cross-country comparable data for 27 EU member states, as well as Iceland and Norway, from the European Union Statistics on Income and Living Conditions (EU-SILC) survey.
Video source: UNICEF-IRC: Multidimensional Child Poverty.
I have been engaged with research and teaching in social policy ever since my master programme MSc in Public Policy and Human Development at the Maastricht Graduate School of Governance – UN-MERIT during which I specialized in Social Policy Design (2009-2010).
As part of my master thesis, I carried out a three-months fieldwork in Mauritius evaluating the effectiveness and efficiency of a national-level unemployment reduction programme called ‘Placement for Training’. Since then, I have been involved in multiple research projects regarding social policy. Whilst working at the UNICEF Office of Research
(2011-2014), some of my tasks included mapping social protection programmes, carrying out cost analyses of social assistance, and doing background research on family policy in Europe and cash transfer programmes in Africa. Whilst working as a social policy consultant at the Economic Policy Research Institute
(2015-2016), one of my main projects involved mapping social protection programmes in Morocco, analysing their coverage of the various risk groups, and developing a vision for a more integrated social protection system. It was carried out with the support of the Ministry of General Affairs and Governance in Morocco and UNICEF. During this project, I met various counterparts of different ministries in Morocco which allowed me to acquire an understanding on how the different ministries cooperate and interact, and how important cooperation and convergence between the different counterparts can be in shaping policies and reaching the people in need.
Another type of experience which has allowed me to progress in the field of social policy is teaching
. I have designed and lectured a course in ‘Quantitative Social Protection Analysis’ as part of an MSc programme offered at the University of Mauritius
to social policy professionals from governmental institutions in developing countries. I have taught this two-week course for six consecutive years (2011-2016), allowing me to gain teaching experience and to acquire more in-depth knowledge about the social protection systems of my students’ home countries.
Geographically, my research focuses mainly on low- and middle-income countries, but I also have an interest and experience in analysing social policy in high-income countries. As an example, I contributed to one of the background papers of the UNICEF Report Card 12 related to child well-being and changes in family policy during the latest recession in rich countries. Currently I am analysing social investment policy returns in the OECD countries together with Prof. Anton Hemerijck
at the European University Institute (EUI)
. At a personal level, reviewing changes in family-related social policies and their impacts in Latvia and across countries is particularly engaging. I have given presentations on this topic
during conferences organized by the cabinet of the President of Latvia, published opinion articles
in Latvian newspapers, and presented at international conferences
such as ESPANet – The European Network for Social Policy Analysis
Research in Progress:
The Social Investment Litmus Test: Family Formation, Employment and Poverty. With Anton Hemerijck. Revise & Resubmit for the European Journal of Social Policy.
Keywords: Social Investment returns; Life-course; Families with children; Employment; Poverty; OECD
Over the past decade, the notion of ‘social investment’ has gained considerable traction in the policy debate over welfare state futures, especially in Europe. Social research is swiftly following suit. We argue in this contribution that many social science attempts fail to come to grips with the new reality of social investment reform and associated life outcomes. Like any notion of investment, social investment posits social returns on investment. Fundamental to any empirical assessment of social investment performance is the acknowledgement of its dynamic mid- to long-term timeframe in a life-course perspective. This paper contributes to improving the assessment and measurement of social investment policy returns over the life-cycle, with a special focus on the period of family formation. It opens by setting the scene of the effects of social investment policies and how they are commonly measured, explicating the main drawbacks in the currently applied methods of measurement. The paper conceptualises the effect of social investment policies in terms of people’s ability to meet their needs with regards to employment, economic independence, and living standards. We use macro-level data across countries in the OECD area to analyse the longer-term cumulative effects of social investments in ‘stocks’, ‘flows’ and ‘buffers’ for one of the most important stages in the life-cycle: that of transitioning into employment and family formation.
How to explain policy recalibration and inertia? The case of post-1989 family policy in Latvia and Poland. With Maciej Sobocinski.
Paper for the ESPANet 2018 Conference.
Over the past decade, the notion of ‘Social Investment’ (SI) has gained considerable traction in the political debates over welfare state futures. The multifaceted character of SI policy interventions, the effects of policy complementarities and interactions for different social groups and generational cohorts, and the challenge of delineating effects across different time dimensions, we argue, are not (yet) properly addressed by current empirical research. This paper contributes to reorienting the measurement of SI returns into a longer-term perspective, conceptualising them as people’s work- and welfare- related outcomes. It operationalizes in a novel fashion macro-level data across OECD countries to analyse the medium-term aggregate effects of SI stock, flow and buffer policies with a focus on arguably the most critical stages in the post-industrial life-cycle course: transition into employment and family formation. Our findings imply that the so-called Matthew Effects identified in previous research stem from a measurement of SI returns conceptualized in a short-term redistributive perspective. Moving on to longer-term returns to SI policies at the societal level reveals positive outcomes for families with children.
Mapping of the current social protection system in Morocco – Mapping de la Protection Sociale au Maroc. With Aminata Bakouan Traoré, Virginie Reboul, Chris de Neubourg. 2018, Ministère Délégué auprès du Chef du Gouvernement chargé des Affaires Générales et de la Gouvernance (MAGG), UNICEF, Nations Unies Maroc.
Comme dans la plupart des pays du monde, les composantes du système de protection sociale au Maroc ont été développées de manière fragmentée sur une période très longue. Aujourd'hui le système au Maroc se compose d’un système de sécurité sociale qui est contributif (CMR, RCAR, CNSS, CNOPS, …), d’un système de protection sociale partiellement contributif (RAMED), et d’un système de protection sociale non-contributif (Tayssir, Kafala, IINDH, établissements de protection sociale pour les personnes en difficulté. Des gains d'efficience et des gains d'efficacité peuvent être réalisés par une reformulation des composantes de la protection sociale marocaine dans un système plus intégré ; l'efficience croît lorsque le même résultat est délivré à moindre coût; l’efficacité est améliorée lorsque les résultats escomptés sont obtenus pour les familles, les enfants, les plus vulnérables et les bénéficiaires en général. Cette étude vise à réorganiser et à réformer les éléments existants dans le contexte de la réalité sociale, économique et politique du Maroc et proposer d’autres composantes destinées aux catégories sociales non couvertes. Dans l'étude, une attention particulière est accordée à la situation des enfants dans la société marocaine selon une approche axée sur l’équité. L’étude a donc pour objectif global l'élaboration d'une vision intégrée et harmonisée de la protection sociale au Maroc, à partir de l’analyse de l'état des lieux et à la lumière des bonnes pratiques internationales dans ce domaine.
En mars 2015, un atelier de lancement du processus d’élaboration d’une vision intégrée de la protection sociale au Maroc a été organisé par le Ministère Délégué auprès du Chef du Gouvernement Chargé des Affaires Générales et de la Gouvernance (MAGG) avec l’appui de l’UNICEF. L’atelier, qui était la première étape du processus, réunissait les principaux départements ministériels et organismes impliqués dans la protection sociale dans le Royaume. Cet atelier a été suivi en avril de la même année, d’une série d’entretiens bilatéraux menés par l’équipe de consultants avec les responsables de la formulation et de la mise en œuvre des composantes de la protection sociale existantes. Les entretiens ont permis à l’équipe d’obtenir des informations très détaillées sur les programmes en cours. Le présent rapport, la cartographie du système de protection sociale au Maroc, a été élaboré sur la base de ces entretiens bilatéraux et d’une revue documentaire détaillée. Dans ce rapport, l’étude fait une analyse à trois niveaux : une analyse de l’intégration au niveau politique (analyse du système: objectifs et fonctions bien définis et accordés), une analyse de l’intégration au niveau programmatique (l’équilibre entre les composantes du système, les chevauchements, les duplications ou contradictions; les composantes harmonisées, pertinentes et consistantes) et enfin une analyse de l’intégration au niveau de la gouvernance et de l’administration (les bases de données coordonnées, les procédures harmonisés, les systèmes de suivi et évaluation).
The Cost of Social Cash Transfer Programs in Sub-Saharan Africa. With M. de Milliano and S. Handa. 2014, The Transfer Project: Research Brief.
Photos from personal archives