Top Banner
CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan Africa Marlous de Milliano and Ilze Plavgo Office of Research Working Paper WP-2014-19 | November 2014
41

CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

Jan 12, 2019

Download

Documents

lamthu
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

1

CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis:

Analysing Child Poverty and Deprivation in sub-Saharan Africa

Marlous de Milliano and Ilze Plavgo

Office of Research Working Paper

WP-2014-19 | November 2014

Page 2: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

2

INNOCENTI WORKING PAPERS

UNICEF Office of Research Working Papers are intended to disseminate initial research

contributions within the programme of work, addressing social, economic and institutional aspects

of the realization of the human rights of children.

The findings, interpretations and conclusions expressed in this paper are those of the authors and

do not necessarily reflect the policies or views of UNICEF.

This paper has been extensively peer reviewed both internally and externally.

The text has not been edited to official publications standards and UNICEF accepts no responsibility

for errors.

Extracts from this publication may be freely reproduced with due acknowledgement. Requests to

utilize larger portions or the full publication should be addressed to the Communication Unit at

[email protected].

For readers wishing to cite this document we suggest the following form:

de Milliano, M. and I. Plavgo (2014). Analysing Child poverty and deprivation in sub-Saharan Africa:

CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis, Innocenti Working Paper

No.2014-19, UNICEF Office of Research, Florence.

© 2014 United Nations Children’s Fund (UNICEF)

ISSN: 1014-7837

Page 3: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

3

THE UNICEF OFFICE OF RESEARCH

In 1988 the United Nations Children’s Fund (UNICEF) established a research centre to support its

advocacy for children worldwide and to identify and research current and future areas of UNICEF’s

work. The prime objectives of the Office of Research are to improve international understanding

of issues relating to children’s rights and to help facilitate full implementation of the Convention

on the Rights of the Child in developing, middle-income and industrialized countries.

The Office aims to set out a comprehensive framework for research and knowledge within the

organization, in support of its global programmes and policies. Through strengthening research

partnerships with leading academic institutions and development networks in both the North and

South, the Office seeks to leverage additional resources and influence in support of efforts

towards policy reform in favour of children.

Publications produced by the Office are contributions to a global debate on children and child

rights issues and include a wide range of opinions. For that reason, some publications may not

necessarily reflect UNICEF policies or approaches on some topics. The views expressed are those

of the authors and/or editors and are published in order to stimulate further dialogue on child

rights.

The Office collaborates with its host institution in Florence, the Istituto degli Innocenti, in selected

areas of work. Core funding is provided by the Government of Italy, while financial support for

specific projects is also provided by other governments, international institutions and private

sources, including UNICEF National Committees.

Extracts from this publication may be freely reproduced with due acknowledgement. Requests to

translate the publication in its entirety should be addressed to: Communications Unit,

[email protected].

For further information and to download or order this and other publications, please visit the

website at www.unicef-irc.org.

Correspondence should be addressed to:

UNICEF Office of Research - Innocenti

Piazza SS. Annunziata, 12

50122 Florence, Italy

Tel: (+39) 055 20 330

Fax: (+39) 055 2033 220

[email protected]

www.unicef-irc.org

Page 4: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

4

ANALYSING CHILD POVERTY AND DEPRIVATION IN SUB-SAHARAN AFRICA: CC-MODA – CROSS COUNTRY MULTIPLE OVERLAPPING DEPRIVATION ANALYSIS Marlous de Milliano and Ilze Plavgo UNICEF Office of Research, University of Florence

[email protected] and [email protected]

Abstract. This paper analyses multidimensional child deprivation across thirty countries in sub-

Saharan Africa, applying the Multiple Overlapping Deprivation Analysis (MODA) methodology that

measures various aspects of child poverty. The methodology has been adapted to the particular

needs of this cross-country comparative study, standardising the indicators and thresholds to allow

comparability across countries. Child poverty is defined as non-fulfilment of children’s rights to

survival, development, protection and participation, anchored in the Convention on the Rights of

the Child. DHS and MICS household survey data is used, taking the child as unit of analysis and

applying a life-cycle approach when selecting dimensions and indicators to capture the different

deprivations children experience at different stages of their life. The main objective of the paper is

to present a direct method of child poverty measurement analysing deprivations experienced by

the child. The paper goes beyond mere deprivation rates and identifies the depth of child poverty

by analysing the extent to which the different deprivations are experienced simultaneously. The

analysis is done across thirty countries in sub-Saharan Africa that together represent 78% of the

region’s total population. The findings show that 67% of all the children in the thirty countries

suffer from two to five deprivations crucial to their survival and development, corresponding to

247 million out of a total of 368 million children below the age of 18 living in these thirty countries.

For the other 15 countries of sub-Saharan Africa where the CC-MODA analysis could not be carried

out, predictions of child deprivation rates have been made using GDP per capita, urban population

share, and population size. Based on the actual as well as the predicted multidimensional

deprivation rates, just under 300 million children in sub-Saharan Africa are multidimensionally

poor, being deprived in two to five dimensions crucial for their survival and development. The

findings are also compared with other existing poverty measures, showing that for the countries

included in the analysis, monetary poverty measures (both the international $1.25 a day and

national poverty measures) are weak predictors of multidimensional child poverty. The study finds

stronger correlation between multidimensional child deprivation and GDP per capita. The paper

underlines that monetary poverty and multidimensional deprivation are conceptually different,

complementary poverty measures and that there are advantages in measuring both

simultaneously, especially when measuring child poverty.

Keywords: child poverty; multidimensional poverty and deprivation; child rights

Acknowledgements: We are grateful for the valuable contribution of many UNICEF colleagues, as

well as the researchers working on multidimensional poverty measurement in OPHI, the University

of Bristol, the University of Maastricht, and the University of Sussex, for their advice and

inspiration. We are especially thankful to Chris de Neubourg, Jingqing Chai, Ziru Wei, Sudhanshu

Handa, and Goran Holmqvist for their substantive engagement throughout the project. We are also

very grateful to Keetie Roelen and Pierre Martel for their valuable comments and suggestions

when reviewing this paper.

Page 5: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

5

TABLE OF CONTENTS

1. Introduction 6

2. Background 7

3. Methodology 8

4. Findings 11

4.1 Single Deprivation Analysis 12

4.2 Deprivation Count for each Child: Deprivation distribution within and across countries 14

4.3 Deprivation Overlap Analysis 16

4.4 Multidimensional Deprivation Ratios 18

4.5 Decomposition of the Adjusted Multidimensional Deprivation Headcount 21

4.6 GDP per capita and Multidimensional Child Deprivation 22

4.7 Monetary Poverty and Multidimensional Child Deprivation 23

4.8 Multidimensional Deprivation among Children in sub-Saharan Africa 27

5. Conclusion 29

References 31

Annexes 33

Page 6: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

6

1. INTRODUCTION

This paper brings together the results of multidimensional deprivation analyses for thirty countries

in sub-Saharan Africa.1 As these thirty countries represent 78% of the total population in the

region, the paper also tries to shed light on the incidence and depth of child poverty across sub-

Saharan Africa as a whole. The analysis is child-centred taking into account various aspects of

children’s well-being, anchored in the Convention on the Rights of the Child. The paper presents

key results on single and multiple deprivations of children within and across the selected countries.

The results summarise part of larger-scale research using UNICEF’s Multiple Overlapping

Deprivation Analysis (MODA) methodology that extends to the existing approaches in the field of

child poverty. For this study, child poverty is defined as the non-fulfilment of children’s rights to

survival, development, protection and participation at different stages of children’s life.

More than a decade of research using multidimensional poverty measures has brought significant

progress in improving the measurement and understanding of poverty. Alongside monetary

poverty measurement, it has become more and more common to analyse the various direct

deprivations that people experience, and to analyse such deprivations simultaneously. Research

analysing the material well-being and monetary poverty simultaneously has found strong, but far

from complete, correlation between the two (e.g., Perry, 2002; Roelen and Notten, 2011; Roelen et

al, 2011; Bradshaw et al., 2008; Nolan and Whelan, 2011, among others), highlighting the

usefulness of regarding these as two distinct aspects of poverty. The identification of the monetary

poor individuals and those who are deprived of the basic goods and services necessary for their

survival and development can lead to better understanding of the situation people are faced with,

and therefore to better targeted and more effective policy responses. Similarly, it has been

recognised that it is necessary to make a distinction between household poverty and child poverty,

acknowledging that children may experience poverty differently to adults and that people’s needs

differ depending on their age.

Building on the existing methodologies of measuring poverty and as part of UNICEF's continued

efforts to generate quality evidence on child poverty and disparities, UNICEF Office of Research has

developed the Multiple Overlapping Deprivation Analysis (MODA) methodology to measure child

poverty (de Neubourg et al, 2012). This methodology has been created primarily for country-

specific child poverty analyses, where it is applied to specific national contexts with customised

dimensions, thresholds and indicators, utilising the best available household surveys and national

datasets. Although subject to data availability, national MODA analyses generally seek to comprise

both monetary child poverty and child deprivations. The analysis makes a conceptual distinction

between the two measures of poverty and includes a study of their overlap.

Cross-country multidimensional child deprivation analysis (CC-MODA) is a specific application of

the MODA methodology, analysing child deprivation for low- and lower-middle income countries

according to internationally accepted standards of child well-being, using internationally

1 Benin, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Comoros, Republic of Congo, Democratic Republic of Congo, Côte d'Ivoire, Equatorial Guinea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Kenya, Lesotho, Malawi, Mozambique, Niger, Nigeria, Rwanda, Senegal, Sierra Leone, Swaziland, Tanzania, Togo, Uganda, Zimbabwe.

Page 7: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

7

comparable datasets that contain child-specific information. The study builds on the idea of

UNICEF’s Global Study on Child Poverty and Disparities 2007-2008 (Gordon et al., 2003; UNICEF,

2007), enhancing knowledge on child poverty and thus strengthening the position of children in

global discussions on development.

This research paper summarises the CC-MODA methodology and compiles the main results of the

analyses of thirty countries in sub-Saharan Africa. The main purpose of the paper is to present a

methodology that directly measures children’s poverty, and to identify the depth of poverty among

children in sub-Saharan Africa by analysing the extent to which the different deprivations are

experienced simultaneously. The succeeding section gives a more detailed introduction to the

MODA methodology and the specifics of CC-MODA, followed by a presentation of the results of

thirty selected African countries comprising single deprivation analyses, dimensional deprivation

counts, deprivation overlap analyses, multidimensional deprivation ratios, and the comparison

between child deprivation and other existing poverty measures.

2. BACKGROUND

The Convention on the Rights of the Child, ratified by most countries in the world, determines that

children have the right to survival, development, protection and participation (United Nations,

1989). The MODA methodology defines child poverty as non-fulfilment of the rights listed in this

convention, moving from household-level to child-level poverty measurement.2 The approach

permits concentration on the access to various goods and services, as well as freedom from

violence and exploitation, which are all crucial for children’s survival and development. It allows

analysis of the multitude and interrelation of children's deprivations and helps identify the socio-

economically disadvantaged groups of children that experience multiple, overlapping deprivations,

recognising that multiple deprivations have significant adverse effects on individuals, especially

children (See de Neubourg et al., 2012a; 2014 for more background).

MODA methodology has built upon existing approaches of multidimensional poverty

measurement, such as UNICEF's Global Study on Child Poverty and Disparities (see Gordon et al.

2003; UNICEF, 2007), OPHI's Multidimensional Poverty Index (see Alkire and Santos, 2010; Alkire

and Foster, 2011), and other research carried out in the field of multidimensional poverty. The

methodology sets itself apart through combining the following key elements:

It selects the child, rather than the household, as unit of analysis, since poverty may affect

children and adults differently. Regarding children’s developmental needs and the

potential long-term effects, it is beneficial to calculate poverty levels separately for

children;

It emphasises the inclusion of individual-level indicators whenever possible as there may be

differences across children of the same age and children in the same household. Indicators

are only adopted if they are relevant to a child as opposed to indicators which have

2 While many of the indicators are collected at a household level, the analysis is child-centred as the unit of analysis is the child rather than the household. Some of the goods and services are measured at individual level (e.g., vaccinations and education), while others are measured at household level and applied to all children of the same household when relevant to all household members (e.g., access to improved water source or housing material).

Page 8: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

8

relevance to other people in the child’s household but not directly to the child observed.

Household level information is only applied if it has a direct relation with the child’s well-

being;

It adopts a life-cycle approach that reflects the changing needs between early childhood,

primary childhood, and adolescence. By separating the analysis in age-groups, the

methodology facilitates the selection of age-specific indicators, such as school attendance,

timely receipt of vaccinations, etc.;

It applies a whole-child oriented approach by measuring the number of deprivations each

child experiences simultaneously, revealing those most deprived;

It broadens the scope of compartmentalised, sector-based approaches through overlapping

deprivation analyses; and

It generates profiles based on geographical and socio-economic characteristics of the

multiply deprived, allowing for better targeted, more effective policy responses and

interventions.

The cross-country (CC-MODA) analysis has been developed to measure child poverty across

countries by applying international standards as guiding principles for the construction of a core

set of dimensions and indicators that are essential to any child's development irrespective of their

country of residence, socio-economic status, or culture. This particular study presents findings on

child poverty in sub-Saharan Africa using the results of CC-MODA.

3. METHODOLOGY

The findings in this paper are based on the cross-country application of the MODA methodology

(CC-MODA). The details of the general MODA methodology are set out in de Neubourg et al.,

2012a, while the specificities regarding this multi-country study (e.g. choice of datasets, indicators,

dimensions, and data treatment) are given in the CC-MODA Technical Note (de Neubourg et al.,

2012b). Results by country can be found in the web-portal: http://www.unicef-irc.org/MODA/. This

particular paper has required merging together data of thirty selected countries in sub-Saharan

Africa, using data from DHS and MICS surveys carried out between 2008 and 2012 (see Annex 1 for

the exact survey year per country). The results are weighted by the respective child population of

each country.3 The population size by country and age-group can be seen in Annex 2.

CC-MODA uses a rights-based approach to child poverty following the rights covered in the

Convention on the Rights of the Child (CRC) (United Nations, 1989). Table 1 lists the main

dimensions of child well-being retrieved from the CRC and used as the basis for selecting the

different dimensions for measuring child poverty.

3 Child population is calculated by multiplying the percentage of children as a share of the total population of each country (authors’ calculations based on MICS/DHS data) with the total population of each country in 2012 (World Bank, 2014).

Page 9: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

9

Table 1 - Child Well-being Dimensions according to the Convention on the Rights of the Child

Categories Dimensions

Survival and

Development

Food, nutrition; Water, sanitation; Health care; Environment/pollution (CRC Art. 24);

Shelter, housing (CRC Art. 27); Education (CRC Art. 28); Leisure; Cultural activities

(CRC Art. 31); Information (CRC Art.13, 17)

Protection

Exploitation, child labour (CRC Art. 32); other forms of exploitation (CRC Art. 33-36);

Cruelty, violence (CRC Art. 19, 37); Violence at school (CRC Art. 28); Social security

(CRC Art 16, 26, 27)

Participation

Birth registration, nationality (CRC Art. 7, 8); Information (CRC Art.13, 17); Freedom of

expression, views, opinions; Being heard; Freedom of association (CRC Art.12-15).

Source: www.unicef.org/crc (article numbers refer to the Convention on the Rights of the Child, 1989)

The selection of indicators and thresholds for CC-MODA is guided by internationally accepted

standards assuring the relevance to children's development irrespective of their country of

residence, socio-economic status, or culture (details on each of the indicators and thresholds are

given in the CC-MODA Technical Note, De Neubourg et al., 2012b). It uses Demographic and Health

Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) for their relatively rich information on

child deprivations as well as to ensure international comparability of the data.

Since CC-MODA aims to capture child poverty across low- and middle-income countries, the lack of

data availability in some key dimensions of child well-being as well as missing values for certain

groups of children in the DHS and MICS datasets present key challenges to the selection of

dimensions and indicators. The choice of variables for CC-MODA has been largely shaped by data

availability leading to the selection of only eight dimensions of well-being, which are used within

two age-groups. The dimensions of water, sanitation, housing, and protection from violence, refer

to all children irrespective of their age, while nutrition and health are measured for children below

age five,4 and education and information are measured for school-age children and adolescents.

Since the analysis is based on the DHS and MICS surveys, monetary poverty analysis is not included

in this specific application of the MODA methodology.5

Following a life-cycle approach CC-MODA is conducted distinguishing two age-groups: children

below the age of five (infancy and early childhood), and children of age 5 to 17 (school age and

adolescence). The dimensions selected for the CC-MODA analysis are grouped according to these

two life stages with specific indicators relevant to each age-group. Most findings are presented by

age-group, while key outcomes are also presented by combining the two age-groups.

Five or six dimensions are analysed for any one child, depending on the availability of the

'protection from violence' dimension. Although an essential aspect of child protection, this

dimension is not available in the datasets of some of the countries included in the study. To ensure

comparability, the main findings of the cross-country MODA in this paper are based on five

dimensions, thus excluding the violence dimension. Where possible, results based on six

4 Although nutrition and health are crucial for child well-being regardless of age, these dimensions are not included in the analysis for children aged 5-17 due to a lack of adequate information for this age group in the DHS or MICS datasets. 5 Monetary poverty has been included in the National (N-MODA) child poverty and deprivation studies of Senegal, Mali, and Madagascar (see UNICEF Senegal, forthcoming; De Milliano and Handa, forthcoming; Plavgo, forthcoming), due to household consumption/expenditure module availability in the data.

Page 10: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

10

dimensions are also presented (see Annex 1 for information on the countries, datasets and

availability of the protection from violence dimension).

Figure 1 – Life-cycle stages, dimensions and indicators used for the CC-MODA analysis

1 Infant and young child feeding:

breastfeeding and food frequency

1 Compulsory school attendance

2 Wasting 2 Primary school attainment

1 DPT immunization 1 Availability of information devices

1 Access to improved water source 1 Access to improved water source

2 Distance to water source 2 Distance to water source

1 Access to improved sanitation 1 Access to improved sanitation

1 Overcrowding 1 Overcrowding

2 Floor and roof material 2 Floor and roof material

1 Domestic violence 1 Domestic violence

The six deprivation dimensions comprise a total of thirteen indicators, one or two per dimension.

For the dimensions with two indicators, a child is considered deprived if he or she is deprived in at

least one of the two indicators (i.e., the union approach). The rationale for using this approach is

to capture all children showing any sign of deprivation in a specific dimension. If there is more than

one indicator per dimension, the selected indicators have been chosen to complement each other

in the explanation of the dimension they represent and to jointly identify the children’s status in

the respective dimension. This method does not account for the depth of deprivation within a

given dimension because the methodology is developed to focus on the dimensions rather than

separate indicators. The indicators and their thresholds are selected based on the international

standards set by the WHO, UN-HABITAT, UNESCO, and UN MDGs (see De Neubourg et al 2012b, for

the precise definition of each indicator).

CC-MODA comprises various steps to analyse multidimensional deprivation. The analysis starts

with a single deprivation analysis to inform about the deprivation levels in each of the indicators

and dimensions included in the multiple deprivation analysis. The multiple deprivation analysis

comprises the following components: (1) deprivation count and distribution analysis, (2)

deprivation overlap analysis, and (3) multidimensional deprivation ratios and their decomposition.

The single deprivation analysis shows deprivation rates per indicator and dimension and per age-

group. Results are presented as child population shares in which the nominator is the number of

children being deprived in the given variable, and the denominator is all children with information

in the respective indicator or dimension. For the multiple deprivation analysis, the numerator is the

cumulative number of deprivations among children deprived in the selected number of dimensions

(out of a total of five dimensions). From this aggregate the proportion of children deprived in 0, 1,

2, .., 5 dimensions is calculated using all children in the selected age-group as a denominator.

Page 11: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

11

To identify the multidimensionally poor children, MODA uses the multidimensional deprivation

headcount (H), representing the children whose total number of deprivations is equal to or above a

specified cut-off, as a percentage of the respective child population. Although a good indication of

deprivation incidence, the headcount ratio (H) is not sensitive to the breadth of multidimensional

poverty, as it remains unchanged regardless of whether children who are identified as

multidimensionally poor suffer from two to three or four to five deprivations simultaneously. For

this reason, two additional ratios are used in the analysis, applying the Alkire and Foster (2011)

methodology. The average deprivation intensity among the deprived (A) measures the breadth of

multidimensional deprivation. It is calculated using the number of deprivations that the

multidimensionally deprived children encounter, divided by the maximum number of dimensions

considered, showing the average number of deprivations the deprived children experience. The

adjusted multidimensional deprivation headcount (M0), adjusts the deprivation headcount rate by

the intensity of deprivation (i.e., the number of deprivations the multidimensionally deprived

children experience), and is calculated by the following formula:

𝑀0 =∑ 𝑞𝑘×𝑐𝑘

𝑛×𝑑 , with 𝑐𝑘 = 𝐷𝑖 × 𝑦𝑘

Where

k - cut-off point (no. of dimensions a child should be deprived in to be considered multidimensionally poor)

𝑞𝐾 - number of children affected by at least K deprivations;

cK - number of deprivations each multidimensionally deprived child i experiences

n - total number of children

d - total number of dimensions considered per child

Di - number of deprivations each child i experiences

yK - deprivation status of a child i depending on the cut-off point k, with yK =1 if Di ≥k; yK =0 if Di <k.

The number of multidimensionally deprived children is expressed as a share of the total child

population as well as in absolute numbers. The child population per country is calculated

multiplying the total population size in 2012 (retrieved from the World Bank Databank, Oct 2014)

with the percentage of children as a share of the total population of each country (derived by

authors’ calculations based on DHS/MICS data; see Annexes 1 and 2).

4. FINDINGS

The findings are based on the analysis of thirty countries in sub-Saharan Africa for which a recent

MICS or DHS dataset was available. In 2012 these countries represented 78% of the total

population of sub-Saharan Africa (and 10% of the world’s population). Children below age 18

represent 52% of the total population of these countries.6 The countries included in the analysis

have experienced a large population growth over the last decades, and the trend is expected to

continue. Population projections show a doubling of the African population between 2015 and

2050, predicting that 37% of all children under 18 will be found on the African continent by 2050

(UNICEF, 2014). These expected demographic and social changes strengthen the choice of focusing

6 Estimates based on child population shares per country (authors’ calculations using DHS/MICS data), and on the population estimates in 2012 (World Bank, 2014).

Page 12: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

12

on this specific region, in particular by concentrating on the well-being of children, central to the

projected transitions.

The findings presented in this paper serve a twofold purpose. On the one hand the results show

the total deprivation of all children in the selected sub-Saharan African countries, while on the

other hand the findings function as a comparison of children’s deprivation levels between

countries. The deprivation levels are firstly analysed for each dimension separately, followed by

counting the number of deprivations experienced by each child. The multiple deprivation analysis

shows the intensity of poverty and the distribution of deprivations among children, describes how

the different sectoral deprivations overlap, and analyses the multidimensional deprivation

incidence and severity. In particular, multidimensional deprivation ratios are used for comparing

deprivation levels across the selected countries. The paper also shows the correlation between

multidimensional child poverty and GDP per capita as well as monetary poverty, and predicts child

deprivation rates for the sub-Saharan region as a whole based on the findings of the thirty

countries analysed.

4.1 Single deprivation analysis

Single deprivation analysis is the basis for understanding the situation of children in each of the

sectors analysed. Knowledge of the deprivation levels by dimension creates an advantageous

starting point for the multidimensional deprivation analysis to follow.

The results show that the highest deprivation rates out of the eight dimensions studied in this

region are in sanitation (67% for the younger and 66% for the older age-group), protection from

violence (63% for both age-groups), health (56% for the younger age-group) and water (52% for

younger and 51% and older children), followed by housing (44% for both age-groups), nutrition

(40% for the younger age-group), education (35% for children above the compulsory school

starting age), and information (26% for the older age-group) (see Annex 3). As shown in Figure 2,

there is a considerable difference between the overall deprivation levels depending on where

children live: the deprivation rates are considerably higher in rural areas as compared to urban

areas in almost all dimensions, apart from nutrition and protection from violence. For children

below the age of five the main issues are sanitation (78%) and health (64%) in the rural areas, and

protection from violence (67%) and nutrition (41%) in urban areas. Children aged 5 to 17 mainly

experience deprivations in sanitation (77%), water (62%) and protection from violence (61%) in

rural areas, and protection from violence (66%) and sanitation (34%) in urban locations. Also, more

than one third (35%) of school age and adolescent children in sub-Saharan Africa are deprived in

schooling (41% in rural areas and 20% in urban areas). For the deprivation rates by indicator, see

Annex 3 which provides further details on the drivers of deprivation per dimension.

Page 13: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

13

Figure 2: Deprivation headcount rates by dimension and area

Children below age five Children between age 5 and 17

* Indicates statistically significant differences in deprivation rates by area ((p<0.05).

Unlike monetary poverty which is a household-level measure, deprivation analysis provides more

space to measure individual differences (e.g. gender or intra-household differences) when

individual level indicators are used. In the case of CC-MODA, four indicators are at an individual

level allowing for analysis by gender.7 When looking at the children of all thirty countries jointly,

some gender differences are observed (although there are some variations across countries).

Figure 3 shows the deprivation rates for boys and girls regarding wasting, immunisation, school

attendance, and primary school attainment.8 The general trend in sub-Saharan Africa is that for

children below the age of five, the deprivation rate in wasting is higher among boys (9.4% for boys

vs. 7.8% for girls), while no statistically significant gender differences can be observed in terms of

DPT immunisation. With regards to schooling indicators for older children, the percentage of

children not attending school at compulsory school age is significantly higher among girls, while the

percentage of adolescents without primary education is equally high for both boys and girls and

the difference is statistically insignificant (at a 95% level).

7 Analysis by gender using CC-MODA is only possible at indicator level; it is not done for multidimensional deprivation analysis because five out of eight dimensions are constructed using indicators that are applied to all household members. 8 All the other indicators used for this analysis (i.e., indicators on water, sanitation, housing, and violence) apply to all children from the same household; thus, gender differences cannot be measured.

40%64%

62%

78%

52%

62%41%

30%

21%

34%

20%

67%

0%

25%

50%

75%

100%Nutrition

Health*

Water*

Sanitation*

Housing*

Protectionfrom

violence*

Rural Urban

41%

33%

62%

77%

51%

61% 20%

8%

21%

34%

23%

66%

0%

25%

50%

75%

100%

Education*

Information*

Water*

Sanitation*

Housing*

Protectionfrom

violence*

Rural Urban

Page 14: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

14

Figure 3: Deprivation headcount rates by indicator and gender

* Indicates statistically significant differences in deprivation rates by gender ((p<0.05).

4.2 Deprivation count for each child: deprivation distribution within and across countries

The multiple deprivation analysis moves from sector-analysis to child deprivation analysis,

examining how many and what combinations of deprivations each child experiences

simultaneously. It shows: (1) the distribution of the number of deprivations, (2) the deprivation

overlap between dimensions, (3) multidimensional deprivation ratios, (4) the profile of the

multidimensionally deprived and the most vulnerable, and (5) the contribution of each country and

dimension to the total adjusted multidimensional deprivation ratio. The results create an

understanding about the intensity of deprivation and the overlap of certain dimensions. At a

country-level these types of results can be used as a basis to further identify the most vulnerable in

the society. In a cross-country context, these findings are telling about the overall level of

deprivation among children in the region, and the differences between multiple deprivation levels

across countries.

The figures below show that among children below the age of five in the thirty countries analysed,

8.5% (10.2 million out of a total of 119.7 million) are not deprived in any of the five selected

dimensions, while a similar share of children under five years (8% or 9.4 million) are experiencing

all five deprivations simultaneously. More than half of the children under five in these countries

(54% or 64.3 million) are deprived in three to five dimensions. Among children of the older age-

group the breadth of deprivation is relatively lower, with 36% of children between the age 5 and

17 deprived in only one or none of the five dimensions studied. 41% of the older children lack basic

needs in three to five dimensions, representing 102.2 million children out of the 248.2 million

children in this age-group across the thirty countries analysed.

The difference between the two age-groups is driven by their dimension choice: apart from the

three dimensions that are common and relevant to all children, for the younger children critical

dimensions on health and nutrition are included to capture their survival and developmental

rights. These dimensions focusing on issues of survival and development have higher deprivation

rates than the dimensions on development (education) and participation (information) which are

selected for the children of primary school-age and adolescents.

Overall, 67% of all the children in the thirty countries of sub-Saharan Africa experience two to five

deprivations that are seen as non-fulfilment of their rights to survival, development and

participation. This represents 247 million of a total of 368 million children below the age of 18.

50.7%

24.2%

37.3%

7.8%

51.3%

20.9%

37.4%

9.4%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%

Primary school attainment

Compulsory school attendance*

DPT immunisation

Height for weight (wasting)*

Deprivation rate in %

Boy

Girl

Page 15: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

15

Figure 4: Number of deprivations children suffer from, by age-group

Children below age five Children between age 5 and 17

A large variation within the region can be observed in terms of the distribution of deprivations

among children at the country level. For instance, in Rwanda (Figure 5) nearly one fifth of all

children below the age of five (19.3%) in rural areas are not deprived in any of the five dimensions

analysed while this is so only for 1.3% of the rural children in Tanzania. In rural areas in Rwanda

more than two thirds (65%) of the children under the age of five experience only one or two

deprivations at the same time, while more than a half (61%) of the young children in rural Tanzania

suffer from three or four deprivations simultaneously.9

Figure 5: Deprivation distribution in rural areas by country - children under five years old

Rwanda (DHS 2010-11) Tanzania (DHS 2010)

9 In Rwanda, the highest deprivation rates are in the water dimension, while in Tanzania, the main contributors to multiple deprivation are sanitation, followed by water and health issues. See Figure 12 for decomposition by dimension to see which dimensions contribute the most to multiple deprivation by country.

0

10

20

30

40

50

60

0%

5%

10%

15%

20%

25%

0 1 2 3 4 5

Mill

ion

s o

f ch

ildre

n

% o

f ch

ildre

n 0

-4 y

ears

In % In numbers

0

10

20

30

40

50

60

0%

5%

10%

15%

20%

25%

0 1 2 3 4 5

Mill

ion

s o

f ch

ildre

n

% o

f ch

ildre

n 5

-17

yea

rs

In % In numbers

0%

10%

20%

30%

40%

0 1 2 3 4 5% o

f ch

ildre

n 0

-4 y

ears

in

rura

l are

as

Number of deprivations experienced

Rwanda

0%

10%

20%

30%

40%

0 1 2 3 4 5

% o

f ch

ildre

n 0

-4 y

rs in

ru

ral

area

s

Number of deprivations experienced

Tanzania

Page 16: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

16

Large disparities in the deprivation distributions can also be found when breaking the total child

population into sub-groups based on children’s geographic location or children’s and their parents’

socio-economic characteristics. When comparing deprivation distribution for children below five by

households’ exposure to child mortality, the results show that children living in households where

at least one child has died experience a higher number of deprivations. 62% of children from

households where under-five child mortality has been experienced suffer from three to five

deprivations at the same time, versus 52% of children living in households where no child mortality

has been observed.

4.3 Deprivation overlap analysis

A deprivation overlap analysis aims at improving the understanding of the nature and depth of

child deprivation by analysing the combinations of deprivations that children experience

simultaneously. The knowledge derived from studying deprivation overlap can be used to direct

policy mechanisms and help address children’s needs more adequately.

For example, Figure 6 shows whether dimensional deprivations are unique issues or whether the

deprivation in a given dimension is experienced simultaneously with other deprivations. Knowing

whether particular deprivations are stand-alone issues may be useful when identifying possible

entry points for policy interventions. Figure 6 demonstrates the deprivation incidence for three

dimensions, subdivided by the extent of overlap with other dimensions by urban and rural areas

separately. The first part of the bar (on the left) shows the proportion of children deprived in only

the specified dimension and no other dimensions, while the other parts of each bar show how

many children are deprived in the specified dimension and also other dimensions simultaneously.

The graph shows that although the percentage of children deprived in nutrition is similar among

children below the age of five in both, rural and urban areas (the entire bar; 39% for both), the

extent to which children experiencing malnutrition are exposed to also other deprivations differs

depending on where children live. Out of all the children under five in urban areas who are

deprived in nutrition, more than one-third (37%) experience malnutrition as a unique problem,

while this is so only for 6% of the malnourished children in rural areas. More than a half (58%) of

the malnourished children living in rural areas are deprived in three to five other dimensions, while

this is so only for 10% of the malnourished children living in urban areas.

Similarly, health and education cannot be regarded in isolation from other deprivations as they are

not stand-alone problems in the region. Most of the children who experience these deprivations

suffer also from other deprivations (Figure 6). In rural areas, for example, roughly one half of all the

children deprived of access to health care or deprived of education, experience three to five other

deprivations.

Page 17: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

17

Figure 6: Deprivation overlap by dimension

The results from the overlap analysis reconfirm the need for integrated approaches to address the

multiple facets of children’s poverty. For instance, Figure 7 shows that in rural areas, while 39% of

all children below the age of five are deprived in nutrition (the pink circle), 90% of these

malnourished children also suffer from lack of access to health facilities or services and/or are

using an unimproved toilet or latrine. Thus, addressing the nutritional problems of these children

would only solve one of the many problems they are faced with; even if their nutritional status was

improved, they would still suffer from other deprivations crucial to their survival. A holistic

approach to resolve children’s problems in an integrated manner would in this case be more

efficient and effective in safeguarding children’s rights to survival and development. Identifying the

children suffering from single and multiple deprivations can help to target the interventions. For

instance, Figure 7 also shows that the overlap of nutrition, health, and sanitation deprivations is

much smaller for children living in urban areas, which may need different intervention strategies

than for children in rural areas where more than one fifth of all children experience all three

specified deprivations at the same time.

Figure 7: Overlap of nutrition, health and sanitation by area of residence

Rural areas Urban areas

0% 10% 20% 30% 40% 50% 60% 70%

Rural

Urban

Rural

Urban

Rural

Urban

Edu

cati

on

(5-1

7)

Hea

lth

(0

-4)

Nu

trit

ion

(0

-4)

Deprived only in the specified dimension Deprived in 1 other dimension

Deprived in 2 other dimensions Deprived in 3-5 other dimensions

Page 18: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

18

4.4 Multidimensional deprivation ratios

To compare the incidence and depth of children’s deprivation across countries and to help

determine the multidimensional poverty threshold, we first identify all children deprived in any

one dimension, and then look at the average number of deprivations these children experience. In

the thirty countries included in the analysis, 86.4% of all the children below the age of 18

experience at least one out of a total of five deprivations analysed (representing 317.7 million

children). These 86.4% of all children on average suffer from 2.6 deprivations simultaneously, as

shown in Figure 8 by the dots representing average deprivation intensity among children deprived

in one to five dimensions. The average number of deprivations among children with at least one

deprivation ranges from 1.7 in Gabon and Swaziland to 3.4 in Chad and Ethiopia. The average

deprivation intensity is calculated using a cut-off of one dimension to avoid censoring the

deprivations that may be experienced in isolation from other deprivations. See Annex 5 for the

percentage of children deprived in one to five dimensions for which the deprivation intensity was

calculated.

For the purposes of this study, children are identified as multidimensionally deprived when

experiencing two or more deprivations out of a total of five dimensions studied (results for all

indices using all possible cut-off points are shown in Annex 5). The results are presented in Figure 8

by the bars showing that the multidimensional deprivation incidence in the selected countries of

sub-Saharan Africa is 67%. In other words, two thirds of all children in this region experience two to

five deprivations, which in absolute numbers is 247 million children. The prevalence of

multidimensional deprivation ranges from 30% in Gabon to 90% in Ethiopia.

This measure shows slightly different results in terms of country ranking when compared to the

measure of deprivation intensity. For example, while the average number of deprivations that

children deprived in any one dimension experience is lower in Malawi compared to Mozambique

(2.6 vs. 2.9 deprivations respectively), Malawi has a higher percentage of children deprived in two

to five dimensions than Mozambique (79% vs. 75%). The combination of these findings means that

the depth of the deprivation is larger in Mozambique, while proportionally Malawi has a slightly

higher share of children deprived in a multitude of dimensions (this holds as well when using a cut-

off of one deprivation; See Annex 5).

Page 19: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

19

Figure 8: Multidimensional deprivation incidence and average deprivation intensity by country for children below 18 years of age

Since the dimensions analysed differ depending on the age of the child, we also look at the

multidimensional deprivation rates for the two age-groups separately. As can be seen in Figure 9,

children below the age of five have a considerably higher multidimensional deprivation incidence

compared to the older children across all countries apart from Malawi. Among children of the

thirty countries analysed, 75% of the children below the age of five experience two to five

deprivations, compared to 64% among school age children and adolescents. The difference

between the deprivation levels of the two age-groups differs by country, which also means that the

ranking of countries by deprivation rate changes based on the age-group chosen. Countries such as

Malawi, Burkina Faso, Benin, and Zimbabwe rank higher when looking at multidimensional

deprivation rates of children aged 5 to17 compared to the children below the age of five. Gambia,

Guinea and Gabon, on the contrary, are performing relatively better with regards to the

deprivation rates for the children in the older age-group. The discrepancy between the deprivation

rates for younger and older children depends on the deprivation levels of the age-specific

dimensions, i.e. nutrition and health for the first age-group and information and education for the

second.

29.6%

67.1%

90.1%

1.7

2.6

3.4

0.00.51.01.52.02.53.03.54.04.55.0

0%10%20%30%40%50%60%70%80%90%

100%G

abo

n

Gam

bia

Swaz

ilan

d

Rw

and

a

Gh

ana

Sen

ega

l

Co

mo

ros

Equ

ato

rial

Gu

inea

Zim

bab

we

Co

ngo

Cam

ero

on

Co

te d

'Ivo

ire

Nig

eria

Ben

in

Leso

tho

Gu

ine

a

Togo

Sier

ra L

eon

e

Ken

ya

Bu

run

di

Bu

rkin

a Fa

so

TOTA

L

Cen

tral

Afr

ican

Rep

.

Uga

nd

a

Mo

zam

biq

ue

Tan

zan

ia

Mal

awi

Co

ngo

DR

Nig

er

Ch

ad

Eth

iop

ia

Ave

rage

dep

riva

tio

n in

ten

sity

(1

-5 d

epri

vati

on

s)

Dep

riva

tio

n r

ate

(2-5

dep

riva

tio

ns)

, as

% o

f al

l ch

ildre

n

% of children deprived in 2-5 dimensions

Average number of deprivations among children with 1-5 deprivations

Page 20: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

20

Figure 9: Multidimensional deprivation rates of children with 2-5 deprivations by age group

The adjusted deprivation headcount (M0) combines the two aforementioned deprivation measures

to show an overall multidimensional deprivation measure that captures both the incidence of the

deprived children and the depth of their deprivation. This ratio ranges between 0 and 1, with zero

showing no deprivation (according to the cut-off chosen) and one showing that everyone included

in the analysis is deprived in all the dimensions analysed. In the thirty countries of sub-Saharan

Africa, the adjusted multidimensional deprivation ratio is 0.42 when using a threshold of two

deprivations (i.e., children are multidimensionally deprived if they suffer from two to five

deprivations). For children under the age of five, the ratio is 0.48, ranging from 0.19 in Swaziland to

0.70 in Ethiopia. For the children of age 5 to 17 years, the total is 0.39 when using the same

threshold, varying between 0.11 in Gabon to 0.62 in Ethiopia (see Annex 5).

Figure 10 presents the findings for all children below age 18 in a map, indicating a few clear

groupings of countries with different deprivation levels. The highest multidimensional deprivation

levels are found at the centre of the continent (Chad, the Democratic Republic of Congo, Niger, and

Central African Republic, ranging between 0.64 and 0.48), followed by a stretch of countries with

high levels of deprivation in the East (Mozambique, Malawi, Tanzania, Uganda and Kenya, ranging

from 0.49 to 0.37), and Burkina Faso, Sierra Leone, Guinea and Togo in West Africa (0.40 – 0.35).

Figure 11 shows the contribution of each country to the total adjusted multidimensional

deprivation ratio of the selected sub-Saharan African countries (total M0 = 0.42). 10. The largest

contributions come from Ethiopia (20%), Nigeria (17%) and the Democratic Republic of the Congo

(13%). The extent to which each country contributes to the total adjusted deprivation ratio

depends not only on the percentage of multidimensionally deprived children per country or their

deprivation intensity (average number of deprivations experienced simultaneously), but also on

10 Annex 4 presents the contribution of each country to the total multidimensional deprivation headcount ratio (H) and the total adjusted deprivation headcount ratio (Mo) by age-group. The patterns are very similar, with the exception of a few cases. For Ethiopia and the Democratic Republic of Congo, the contribution to the total deprivation level in the region is smaller when using the deprivation ratio, H (17% and 12%) than when using the adjusted deprivation ratio M0 (20% and 13%, respectively). This is because the adjusted deprivation ratio takes into account that the average number of deprivations children in these countries experience is relatively higher compared to other countries with a similar deprivation headcount ratio. For Nigeria, on the other hand, the contribution to the total deprivation level of the selected countries is slightly smaller when using the adjusted deprivation ratio M0 than when using the deprivation headcount ratio H (17% vs. 18%) due to the differences in the average intensity of deprivation across countries.

37%

75%

94%

32%

64%

88%

0%

20%

40%

60%

80%

100%

Swaz

ilan

d

Gab

on

Rw

and

a

Gh

ana

Gam

bia

Sen

ega

l

Zim

bab

we

Co

mo

ros

Co

ngo

Equ

ato

rial

Gu

inea

Cam

ero

on

Co

te d

'Ivo

ire

Ben

in

Nig

eria

Leso

tho

Togo

Ken

ya

Bu

rkin

a Fa

so

Bu

run

di

Sier

ra L

eon

e

Gu

ine

a

TOTA

L

Cen

t. A

fric

an R

ep.

Uga

nd

a

Mal

awi

Mo

zam

biq

ue

Tan

zan

ia

Co

ngo

DR

Nig

er

Ch

ad

Eth

iop

ia

Per

cen

tage

of

child

ren

dep

rive

d

in 2

-5 d

imen

sio

ns

Age 0-4 Age 5-17

Page 21: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

21

the size of the child population per country (see Annex 2). For this reason, countries such as Chad

with a high deprivation incidence and intensity contribute relatively little since their child

population is small compared to the other countries analysed. The composition of the pie chart

helps to understand where the largest shares of the total amount of all deprivations experienced

across the thirty countries are found.

Figure 10: Adjusted multidimensional deprivation ratio: all children, 2-5 deprivations

Figure 11: Contribution of each country to the total adjusted multidimensional deprivation ratio: all children, 2-5 deprivations

Note: See Annex 1 for country names and respective abbreviations

4.5 Decomposition of the adjusted multidimensional deprivation headcount

In addition to producing a comparable deprivation ratio, one of the special features of the adjusted

multidimensional deprivation ratio is that it can be decomposed. This means that the total

adjusted deprivation ratio can be broken down to show the percentage contribution of each single

dimension to the multidimensional measure. When comparing the composition of the adjusted

deprivation headcount of, for instance, Equatorial Guinea and Malawi, the deprivation in

Equatorial Guinea is mainly driven by health and water deprivations, whereas in Malawi sanitation

and housing play a far larger role compared to the other dimensions. Overall, sanitation and health

are the main contributors to the total adjusted deprivation ratio of all children below the age of

five, apart from Benin, Burkina Faso, Comoros, Congo, Congo DR, the Gambia, and Malawi where

either nutrition or housing issues also dominate, and Rwanda where water deprivation has the

highest contribution. The sanitation dimension contributes considerably to the adjusted headcount

of Benin, Burkina Faso, Comoros, Congo, Lesotho, Malawi and Togo (more than 30% for the

younger age-group). For the school-age children and adolescents, the highest contributor to the

total adjusted deprivation ratio is sanitation, followed by water deprivation, apart from Burkina

BEN1.1%

BFA2.3%

BDI1.3%

CMR2.2%

CAF0.8%

TCD3.0%

COM0.1%

COG0.4%

COD13.3%

CIV1.7%

GNQ0.1%

ETH19.9%

GAB0.1%GMB

0.1%GHA1.7%

GIN1.4%

KEN5.2%

LSO0.2%

MWI2.7%

MOZ4.3%

NER3.8%

NGA16.7%

RWA0.7%

SEN1.1%

SLE0.7%

SWZ0.1%

TZA7.6%

TGO0.8%

UGA5.8%

ZWE1.2%

Page 22: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

22

Faso where the second highest contributor is education, and Lesotho, Malawi, and Senegal where

the second highest contributor after sanitation is housing. Education, information, and housing

have the highest variation across the thirty countries of the sub-Saharan region. The contribution

of education deprivation is relatively high (ranging between 17% and 26% of the total

multidimensional poverty ratio) in the following countries: Burkina Faso, Côte d’Ivoire, Gabon,

Gambia, Ghana, Guinea, Senegal, and Swaziland.

Figure 12: Contribution of each dimension to the total adjusted multidimensional deprivation ratio: children experiencing 2-5 deprivations

4.6 GDP per capita and multidimensional child deprivation

This section looks at the correlation between GDP per capita of each country and the

multidimensional child deprivation rates to assess whether in sub-Saharan Africa a higher average

economic activity is correlated with lower child deprivation levels.

As can be seen from Figure 13, the correlation between per capita GDP and the multidimensional

deprivation measure is moderate. For instance, Chad and Senegal in 2012 had a very similar per

capita GDP, while the multidimensional child deprivation rate in Chad was more than double the

rate of Senegal (88% and 44%, respectively). Similarly, while the per capita GDP in the Republic of

0%

20%

40%

60%

80%

100%

TOTA

L

Ben

in

Bu

rkin

a Fa

so

Bu

run

di

Cam

ero

on

Cen

tral

Afr

ican

Rep

.

Ch

ad

Co

mo

ros

Co

ngo

Co

ngo

DR

Co

te d

'Ivo

ire

Equ

ato

rial

Gu

inea

Eth

iop

ia

Gab

on

Gam

bia

Gh

ana

Gu

ine

a

Ken

ya

Leso

tho

Mal

awi

Mo

zam

biq

ue

Nig

er

Nig

eria

Rw

and

a

Sen

ega

l

Sier

ra L

eon

e

Swaz

ilan

d

Tan

zan

ia

Togo

Uga

nd

a

Zim

bab

we

Co

ntr

ibu

tio

n t

o t

he

adju

setd

dep

riva

tio

n h

ead

cou

nt,

in % Housing

Sanitation

Water

Health

Nutrition

Children below age five

0%

20%

40%

60%

80%

100%

TOTA

LB

enin

Bu

rkin

a Fa

soB

uru

nd

iC

amer

oo

nC

entr

al A

fric

an R

ep.

Ch

adC

om

oro

sC

on

goC

on

go D

RC

ote

d'Iv

oir

eEq

uat

ori

al G

uin

eaEt

hio

pia

Gab

on

Gam

bia

Gh

ana

Gu

ine

aK

enya

Leso

tho

Mal

awi

Mo

zam

biq

ue

Nig

erN

iger

iaR

wan

da

Sen

ega

lSi

erra

Leo

ne

Swaz

ilan

dTa

nza

nia

Togo

Uga

nd

aZi

mb

abw

e

Co

ntr

ibu

tio

n t

o t

he

adju

setd

dep

riva

tio

n h

ead

cou

nt,

in % Housing

Sanitation

Water

Information

Education

Children between age 5-17

Page 23: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

23

Congo was more than three times higher than in Zimbabwe, the child deprivation rates in these

two countries were almost identical (50% and 49%, respectively).

The modest correlation between the two measures may be explained by the fact that GDP per

capita measures country’s average economic activity in terms of monetary transactions, but does

not capture distribution of wealth and societal behaviour. Although the two measures are

correlated to some extent (R2=0.29), 11 the level of the average economic activity of the country is

not a perfect predictor of the level of multidimensional child deprivation. For a more accurate

prediction, other factors should also be taken into account, such as resource use and distribution in

the society, availability and affordability of public and private goods and services, legislation and

legislative accountability, as well as societal behaviours, beliefs and traditions, among others.

Figure 13: GDP per capita and multidimensional child deprivation

4.7 Monetary poverty and multidimensional child deprivation

Following the conceptual framework set out in the MODA methodology, MODA distinguishes two

main concepts of poverty: monetary poverty and multidimensional deprivation (see de Neubourg

et al., 2014, for more details), and uses both to analyse child poverty whenever the data allows.12

Monetary poverty measures the lack of financial means of households to provide the household

members with basic goods and services deemed to be necessary for their survival and

development. Deprivations measure the individual deprivation status in each of the various sectors

considered as crucial for individuals’ survival and development. Deprivations can stem from the

lack of financial means (i.e., monetary poverty), but they can also be the result of unavailability of

basic goods in the market, the lack of service provision, or societal beliefs, customs and behaviour,

among other reasons. Especially regarding children some differences between deprivation and

monetary poverty are expected, as children need goods and services that are more likely to be

subject to missing or incomplete markets (e.g. health care, school or nutritional needs). It is also

11 Gabon and Equatorial Guinea have been omitted as outliers. The regression has also been performed omitting Nigeria, Congo and Swaziland, which gives a considerably stronger correlation (R2=0.51). 12 See footnote 5 for MODA analyses where monetary poverty is measured alongside multidimensional deprivation.

BEN

BFABDI

CMR

CAF

TCD

COMCOG

COD

CIV

ETH

GMB

GHA

GINKEN

LSO

MWI

MOZ

NEG

NGA

RWA

SEN

SLE

SWZ

TZA

TGO

UGA

ZWE

20

40

60

80

10

0

% o

f ch

ildre

n b

elo

w 1

8 d

epri

ved

in 2

-5 d

imen

sio

ns

0 2000 4000 6000GDP per capita PPP in current international USD in 2012

R-squared=0.2882

Page 24: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

24

important to remark that while monetary poverty measurement concentrates on the average

financial means available to the households where children live, deprivation measurement

attempts to determine whether children’s basic needs are satisfied. Measuring child outcomes,

child-related practices, the fulfilment of children’s basic rights gives the possibility to account for

intra-household differences in the distribution of resources; or to reflect decisions (either explicit

or implicitly made) on issues such as schooling, labour and marriage which may be driven by socio-

cultural norms, traditions or lack of awareness rather than lack of resources (see also de Neubourg

et al., 2014, Gordon et al., 2003; Minujín et al., 2006; Minujín and Nandy, 2012). Despite the

expected differences, the MODA methodology encourages analysing both concepts of poverty and

studying the overlap between children experiencing deprivations and children living in monetary

poor families when possible.

Even though the specificities of the CC-MODA analysis do not allow the measurement of monetary

poverty and child deprivation at the same time, we still try to identify the extent to which the two

measures correlate by using aggregate data on international and national poverty rates coming from

other data sources. We look at the correlation between monetary poverty and child deprivation by

using the following two monetary poverty measures: an internationally comparable monetary

poverty rate based on $1.25 PPP a day poverty line, and a nationally determined poverty rate based

on national poverty lines. For a sub-set of countries, we also compare multidimensional child

deprivation rates with child poverty rates based on national poverty lines calculated specifically for

children, so that the same unit of analysis for both measures can be used in the comparison.

Figure 14 compares multidimensional deprivation rates for children with absolute monetary

poverty rates based on the $1.25 PPP per day poverty line for the total population in each country.

We can see a fairly large spread among the countries included in the figure indicating a moderate

correlation between multidimensional deprivation for children and absolute monetary poverty for

the total population. For most of the countries analysed, monetary poverty rates for the total

population are considerably lower than multidimensional deprivation rates for children. There are,

however, a few countries, such as Rwanda and Burundi, with higher monetary poverty levels

compared to deprivation. A sizable group of countries clusters around the trend line suggesting

moderate correlation between the two measures. Nevertheless, the relatively high margin of

unexplained variance between the two measures (R2=0.19) suggests that the absolute monetary

poverty measure is not a good predictor of child deprivation rates in this region.

Page 25: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

25

Figure 14: Monetary poverty based on $1.25 PPP poverty line and multidimensional child deprivation (2-5 dimensions) for all children

Note: Poverty rates retrieved from the World Bank (Oct 2014); see Annex 1 for the poverty rate estimation year per country.

Figure 15 shows national poverty rates that are calculated based on the national poverty line which

is the estimated budget needed to pay for a basic basket of goods and services within a given

national context. These rates are not internationally comparable as different methodologies and

poverty lines have been used to estimate each of the national poverty rates. However, the national

poverty rates are normally used by the governments when identifying the poor and designing

policy responses, making it an interesting measure to be compared with the child deprivation rates

of the respective countries.

As shown in Figure 15, in nine out of thirty countries analysed, monetary poverty rates using

national poverty line and child multidimensional deprivation levels are similar: the two measures

give almost identical poverty and deprivation rates in Burundi (67% and 66%, respectively) and

Lesotho (57% and 58%), and very similar monetary poverty and deprivation rates in Comoros,

Gabon, Republic of Congo, Rwanda, Senegal and Togo. For the remaining countries in sub-Saharan

Africa, however, the poverty and deprivation rates differ considerably. The low R-squared and the

horizontal trend line indicate that there is no correlation between the two measures of poverty

across the thirty countries in sub-Saharan Africa. This finding underlines the usefulness of using

these two measures of poverty in a complementary manner to identify monetary poor and

deprived households and children.

BEN

BFA BDI

CMR

CAF

TCD

COMCOG

COD

CIV

ETH

GABGMB

GHA

GINKEN

LSO

MWI

MOZ

NEG

NGA

RWA

SEN

SLE

SWZ

TZA

TGO

UGA

20

40

60

80

10

0

Perc

enta

ge o

f ch

ildre

n d

ep

rive

d in

2-5

dim

ensio

ns

0 20 40 60 80Poverty rate at 1.25 USD (PPP) a day (% of population)

R-squared=0.1910

Page 26: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

26

Figure 15: Monetary poverty based on national poverty line and multidimensional child deprivation (2-5 dimensions) for all children

Note: Poverty rates retrieved from the World Bank (Oct, 2014); see Annex 1 for the poverty rate estimation year per country.

For a sub-set of the countries included in the sample a monetary child poverty rate is available

providing an opportunity to make a comparison between deprivation and monetary poverty rates

representing the same population. In Ghana and Togo, the difference between the child monetary

poverty and child deprivation rates is the smallest (2 percentage point (p.p.) difference), followed

by Comoros (3 p.p.), and Cameroon, Nigeria and Senegal (5 to 6 p.p.). For Swaziland and

Zimbabwe, the child monetary poverty rates are considerably higher than child deprivation levels.

In the remaining countries for which child monetary poverty rates were available (Benin, Chad,

Côte d’Ivoire, Malawi, Niger and Uganda), the percentage of children deprived in two to five

dimensions is higher than the percentage of children living below the national poverty line. In

general, even though the results on both poverty measures are closer than when using national

poverty rates for the total population, national child poverty rates are not good predictors of the

level of child deprivation and the correlation between the two measures remains weak.

BEN

BFA BDI

CMR

CAF

TCD

COMCOG

COD

CIV

GNQ

ETH

GABGMB

GHA

GINKEN

LSO

MWI

MOZ

NEG

NGA

RWA

SEN

SLE

SWZ

TZA

TGO

UGA

ZWE

20

40

60

80

10

0

Perc

enta

ge o

f ch

ildre

n d

ep

rive

d in

2-5

dim

ensio

ns

20 40 60 80Poverty rate using national poverty line (% of population)

R-squared=0.0006

Page 27: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

27

Figure 16: Child monetary poverty based on national poverty line and multidimensional child deprivation (2-5 dimensions) for all children

Note: Poverty rates retrieved from UNICEF DPR (Sept. 2014); see Annex 1 for the poverty rate estimation year per country.

Each of the above figures shows that the two approaches for measuring child poverty are

complementary to each other as the two measures cannot substitute (or predict) the other. As

argued in de Neubourg et al. (2014) having enough financial resources does not always mean that

the fulfilment of children’s rights is guaranteed. The lack of basic goods and services, as well as the

violation of various children’s rights, can stem from lack of services or infrastructure, lack of

information, administrative restrictions, discrimination, and other reasons. At the same time, it

may well be that the access to certain goods and services is guaranteed without the need of the

financial resources at the household level because, for instance, the goods or services are available

for free or are partially subsidised. The figures above confirm that monetary poverty and

multidimensional deprivation are two different concepts that complement each other when

analysing child poverty.

4.8 Multidimensional deprivation among children in sub-Saharan Africa

Based on the results of the thirty selected countries in sub-Saharan Africa, we have predicted the

multidimensional child deprivation rates for the remaining 15 countries to be able to estimate the

total number of multidimensionally poor children in the whole region.13 In order to predict the

deprivation levels for sub-Saharan countries which are not among the thirty selected countries, we

use an OLS regression model estimating the relationship between multidimensional deprivation

rates and the GDP per capita of the countries included in the analysis. Since the correlation

between multidimensional deprivation and GDP per capita has been found to be only moderately

strong (see Figure 13) the regression has been made more robust by including various control

variables. The regression is based on the multidimensional deprivation levels for 2 to 5 dimensions

13 Calculations are made for 45 developing countries in Sub-Saharan Africa as classified by the World Bank, excluding Mauritius, Seychelles, and Somalia, while adding Equatorial Guinea.

BEN

CMR

TCD

COM

CIV

GHA

MWI

NEG

NGA

SEN

SWZ

TGO

UGA

ZWE

20

40

60

80

10

0

Perc

enta

ge o

f ch

ildre

n d

ep

rive

d in

2-5

dim

ensio

ns

30 40 50 60 70 80Child poverty rate using national poverty line (% of children)

R-squared=0.0632

Page 28: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

28

of 28 countries (excluding Equatorial Guinea and Gabon due to outlier values), and the countries’

GDP per capita, the share of urban population and the population size in 201214 (see Table 2). The

regression is weighted by the countries’ population size. The regression provides coefficients which

are used in the following formula to predict the deprivation rates for the missing countries:

Deprived (2-5 dim)=β0 + β1 * GDP per capita+ β2* Share of urban population + β3 * Population + β4*

Population2

The predicted multidimensional deprivation rates are multiplied by the total number of children15

to estimate the total number of children being multidimensionally deprived in the remaining sub-

Saharan African countries.

Table 2: OLS regression on the relationship between multidimensional deprivation (2-5 dimensions) and GDP per capita

VARIABLES Deprived in 2-5

dimensions

GDP per capita -7.54e-05***

0.000022

Share of urban population -0.290*

0.152287

Population 3.84e-09**

0.000000001

Squared population -1.24e-17

9.23e-18

Constant 0.790***

0.068064

Observations 28

R-squared 0.811

The estimates for the region as a whole show that 298 million out of a total of 468 million children

in the 45 countries in sub-Saharan Africa are multidimensionally poor. In other words, 63.6%, or

just below 300 million children in the 45 countries of sub-Saharan Africa are multidimensionally

poor, being deprived in two to five dimensions of basic child rights out of a total of five dimensions

analysed per child.

To compare these figures with the number of children in the region living in monetary poverty, we

use the extreme poverty rates of the total population applied to the child population per country.

Among the children in the 28 selected countries (361 million children in total), 181 million children

(50%) are living below the extreme poverty line of $1.25 a day, and 244 million children (67.5%) are

multidimensionally deprived. It therefore follows that the proportion of children in poverty based

on CC-MODA (i.e., multidimensionally deprived in 2-5 dimensions) is 17 percentage points higher

than the estimated proportion of children in extreme monetary poverty based on $1.25 a day

poverty line. This is similar also when using the predicted deprivation rates for children in the other

countries of sub-Saharan Africa. Out of a total of 41 countries that are home to 453 million

14 The GDP per capita, share of urban population and population in 2012 are retrieved from the World Bank (2014). 15 Based on our calculations using DHS/MICS data for 30 countries in sub-Saharan Africa, children on average represent 52% of the total population across the region. This proportion is used to calculate the absolute number of children for the 15 remaining countries.

Page 29: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

29

children, 215 million children (47%) live in extreme poverty below $1.25 a day, while 288 million

children (64%) are multidimensionally deprived (see Annex 6 for estimates and a list of countries

included).

It should be noted that the number of children in extreme poverty is likely to be an

underestimation as the calculations are based on poverty rates of the total population, while child

poverty rates are generally higher. The World Bank’s report on ‘the State of the Poor’ reveals that

in developing countries children aged 0 to 18 represent 47% of the population living below the

extreme poverty line of $1.25 PPP, while among the non-poor only 33% are children (Olinto et al,

2013).

Given the data restrictions of this study, we are unable to conclude whether the children living

below $1.25 PPP a day are also multidimensionally deprived, or whether the two measures identify

two different groups of children. To analyse whether the same children lacking the goods and

services crucial to their survival and development are also among the extremely poor in monetary

terms, an overlap analysis is necessary using data that comprises information on both deprivations

and household income/consumption. Based on MODA-related research done so far, monetary

poverty and deprivations overlap to some extent, but a large proportion of children are

multidimensionally deprived, but are not considered as monetary poor, and vice versa.16 This

underlines the need to use the two measures of poverty as complementary measures in order to

identify the poor children.

5. CONCLUSION

This paper is based on the cross-country application of the Multiple Overlapping Deprivation

Analysis (MODA) methodology to analyse the poverty status of children in sub-Saharan Africa. The

methodology has been developed by UNICEF to analyse the number and the combinations of

deprivations children experience, moving from sector-by-sector analyses to a child level analysis by

looking at each child’s outcomes or access to various goods and services and exposure to harmful

practices, to determine children’s status of well-being in the various dimensions simultaneously.

The definition of deprivation is rooted in the child-rights framework using the Convention on the

Rights of the Child as its main source to select dimensions relevant to children’s well-being.

The analysis covers 30 out of 48 countries in sub-Saharan Africa, representing 78% of the total

population in this region. Children below the age of eighteen represent more than a half (52%) of

the total population in the region. The findings show that 67% of all the children across the thirty

countries experience at least two out of five deprivations critical to children’s survival and

development. This percentage represents 247 out of a total of 368 million children in the 30

countries. In an effort to predict the number of multidimensionally poor children in the entire sub-

Saharan Africa, a function was created using regression outcomes including GDP per capita,

population size and share of urban population. This function was then used as a predictor of child

16 For example, research in Senegal (UNICEF Senegal, forthcoming) shows that 20% of children are multidimensionally poor (i.e., lacking several goods or services crucial to their survival and development) but living in families above the national monetary poverty line, which means that a monetary poverty measure alone underestimates children’s poverty; at the same time, 15% of the monetary poor children are not deprived in any of the dimensions analysed, indicating that children’s basic rights may have been fulfilled through public service provision and through other channels. Analyses of child monetary and multidimensional poverty in Mali (de Milliano and Handa, forthcoming) and Madagascar (Plavgo, forthcoming) show similar findings and will be available soon.

Page 30: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

30

deprivation rates for the remaining 15 countries that were not included in the analysis. The results

show that 298 out of a total of 468 million children in the 45 countries in sub-Saharan Africa are

multidimensionally poor, being deprived in two to five dimensions of basic child rights.

In order to place deprivation analysis in the wider context of other measures regarding the wealth

of a country and people’s monetary well-being, the findings of the multidimensional deprivation

analysis were compared to the GDP per capita and national and international poverty. The

comparison between GDP per capita and multidimensional child deprivation headcount ratio

shows a negatively sloped correlation indicating that higher GDP per capita is associated with

lower multidimensional deprivation rates. The monetary poverty measures indicate, somewhat

surprisingly, relatively weak correlation with the multidimensional deprivation measure for

children. The poverty rates based on the internationally comparable $1.25 PPP poverty line

highlight a reasonably strong and positive correlation between the level of monetary poverty and

multidimensional child deprivation. There is, however, a large proportion of unexplained variance

between the two measures of poverty. When looking at the correlation between the poverty rates

based on the national poverty lines and multidimensional deprivation, a weak correlation is

observed, regardless of whether the monetary poverty rates are calculated only for children or for

the total population. These findings could be taken forward by further investigating the correlation

between national poverty rates and country-specific multidimensional deprivation analyses

(resulting from an N-MODA application).

Overall, the combined results of this multidimensional deprivation analysis for children using the

MODA methodology has given an indication of the level of multidimensional deprivation across the

sub-Saharan region. It has provided further details on the depth of deprivation and the

simultaneous experience of deprivations. The comparison between multidimensional deprivation

for children and countries’ GDP per capita has shown a moderate correlation indicating some

ability to predict child deprivation on the basis of a country’s economic activity. In addition, there

has been a modest to absent correlation between deprivation and the various monetary poverty

measures emphasising that the two concepts of poverty identify (partly) different groups of people

and that the two measures of poverty should therefore be used complementary to each other,

especially regarding child poverty.

Page 31: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

31

REFERENCES

Alkire, S., Foster, J. (2011). ‘Counting and Multidimensional Poverty Measurements’, Journal of Public Economics, no. 95, pp. 476-487.

Alkire, S., Santos, M. E. (2010). Acute Multidimensional Poverty: A New Index for Developing

Countries, OPHI Working Paper No. 38, University of Oxford. Bradshaw, J., Hoelscher, P., Richardson, D. (2008). Child Well-being in Central and Eastern

European Countries (CEE) and the Commonwealth of Independent States (CIS), Springer Science.

De Milliano, M., Handa, S., (forthcoming). ‘Child Poverty and Deprivation in Mali – the first national

estimates’, Working Paper 2014-X, UNICEF Office of Research, Florence.

De Neubourg, C., M. de Milliano, I. Plavgo, (2014). 'Lost (in) Dimensions: Consolidating progress in multidimensional poverty research', Working Paper 2014-04, UNICEF Office of Research, Florence.

De Neubourg, C., J. Chai, M. de Milliano, I. Plavgo, Z. Wei (2012a). 'Step-by-Step Guidelines to the

Multiple Overlapping Deprivation Analysis (MODA)', Working Paper 2012-10, UNICEF Office of Research, Florence.

De Neubourg, C., J. Chai, M. de Milliano, I. Plavgo, Z. Wei (2012b). 'Cross-country MODA Study:

Multiple Overlapping Deprivation Analysis (MODA) - Technical note', Working Paper 2012-05, UNICEF Office of Research, Florence.

Gordon, D., Nandy, S., Pantazis, C., Pemberton, S., Townsend, P. (2003). The Distribution of Child

Poverty in the Developing World, University of Bristol. Minujín, A., Delamonica, E., Davidziuk, A., Gonzalez, E. D. (2006). “The definition of child poverty: a

discussion of concepts and measurements”. Environment and Urbanization, 18(2), pp. 481-500.

Minujín, A., Nandy, S. (Eds.) (2012). Global Child Poverty and Well-being – Measurement, concepts,

policy and action. The Policy Press. Nolan, B. and Whelan, C.T. (2011). The EU 2020 Poverty Target. Amsterdam, AIAS, GINI Discussion

Paper 19. Olinto, P., Beegle, K., Sobrado, C., Uematsu, H. (2013). The State of the Poor: Where Are The Poor,

Where Is Extreme Poverty Harder to End, and What Is the Current Profile of the World’s Poor? The World Bank, Economic Premise, Poverty Reduction and Economic Management Network, Oct 2013, No. 125.

Perry, B. (2002). ‘The mismatch between income measures and direct outcome measures of Poverty’, in Social Policy Journal of New Zealand, 19, pp. 101–127. Plavgo, I., (forthcoming). ‘Child Monetary and Multidimensional Poverty Analysis in Madagascar’,

Working Paper 2014-X, UNICEF Office of Research, Florence.

Page 32: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

32

Roelen, K., Gassmann, F., de Neubourg, C. (2011). ‘False positives or hidden dimensions: what can monetary and multidimensional measurement tell us about child poverty in Vietnam?’ in International Journal of Social Welfare 2011.

Roelen, K., Notten, G. (2011). The Breadth of Child Poverty in Europe: an Investigation Into Overlap

and Accumulation of Deprivations, Innocenti Working Paper 2011-04, Florence: UNICEF IRC. UNICEF (2007). Global Study on Child Poverty and Disparities 2007-2008: Guide, Division of Policy

and Planning, New York. UNICEF (2014). Generation 2030, Africa. UNICEF Division of Data, Research and Policy, New York. UNICEF Senegal (forthcoming), Actualisation de l’étude sur la pauvreté et les disparités chez les

enfants au Sénégal 2013. UNICEF Senegal country office, Dakar, Senegal. United Nations (1989). Convention on the Rights of the Child, the General Assembly resolution

44/25. World Bank (2014). World Development Indicators. Washington DC: World Bank.

Page 33: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

33

ANNEX 1 - List of countries and data sources

Country

ISO code

Survey year

Data Violence indicator

Poverty ratio at $1.25 a

day (WB): year of

estimation

Poverty ratio at national

poverty line (WB):

year of estimation

Child poverty ratio

at national poverty line:

year of estimation

Benin BEN 2011-12 DHS No 2012 2011 2010 Burkina Faso BFA 2010-11 DHS Yes 2009 2009 N/A Burundi BDI 2010-11 DHS No 2006 2006 N/A Cameroon CMR 2011 DHS No 2007 2007 2007 Cent. African Rep. CAF 2010 MICS Yes 2008 2008 N/A Chad TCD 2010 MICS Yes 2011 2011 2002 Comoros COM 2012 DHS-MICS Yes 2004 2004 - Congo COG 2011-12 DHS No 2011 2011 N/A Congo DR COD 2010 MICS Yes 2006 2005 N/A Côte d'Ivoire CIV 2011-12 DHS Yes 2008 2008 2008 Equatorial Guinea GNQ 2011 DHS No N/A 2006 N/A Ethiopia ETH 2011 DHS No 2011 2011 N/A Gabon GAB 2012 DHS Yes 2005 2005 N/A Gambia GMB 2010-11 MICS Yes 2003 2010 N/A Ghana GHA 2011 MICS Yes 2006 2006 2005 Guinea GIN 2012 DHS-MICS No 2012 2012 N/A Kenya KEN 2008-09 DHS Yes 2005 2005 N/A Lesotho LSO 2009-10 DHS No 2010 2003 N/A Malawi MWI 2010 DHS No 2010 2010 - Mozambique MOZ 2011 DHS No 2009 2009 N/A Niger NEG 2012 DHS Yes 2011 2007 2011 Nigeria NGA 2011 MICS Yes 2011 2012 2012 Rwanda RWA 2010-11 DHS No 2011 2011 N/A Senegal SEN 2010-11 DHS No 2011 2011 2011 Sierra Leone SLE 2010 MICS Yes 2011 2011 N/A Swaziland SWZ 2010 MICS Yes 2010 2009 2009/10 Tanzania TZA 2010 DHS Yes 2012 2012 N/A Togo TGO 2010 MICS Yes 2011 2011 2011 Uganda UGA 2011 DHS No 2009 2009 2009 Zimbabwe ZWE 2011-12 DHS Yes N/A 2011 -

Note: Poverty estimates retrieved from the World Bank Databank (WB Databank, October 2014).

Page 34: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

34

ANNEX 2 – Share and size of child population by country and age-group

Children below age five Children between age 5-17

Total population, 2012 (WB)

As a share of total population per country

In numbers

As % of children <5 in 30 selected countries

As a share of total population per country

In numbers

As % of children 5-17 in 30 selected countries

Benin 10,050,702 16.4% 1,652,855 1.4% 38.0% 3,821,287 1.5% Burkina Faso 16,460,141 18.0% 2,967,184 2.5% 36.3% 5,973,739 2.4% Burundi 9,849,569 18.4% 1,811,019 1.5% 35.3% 3,474,913 1.4% Cameroon 21,699,631 16.7% 3,621,510 3.0% 34.1% 7,391,185 3.0% Cent. African Rep. 4,525,209 19.7% 890,284 0.7% 33.4% 1,509,219 0.6% Chad 12,448,175 20.3% 2,529,921 2.1% 37.4% 4,657,983 1.9% Comoros 717,503 14.1% 101,032 0.1% 33.3% 239,216 0.1% Congo, Republic of 4,337,051 17.4% 753,622 0.6% 31.4% 1,361,984 0.5% Congo DR 65,705,093 18.5% 12,167,197 10.2% 35.6% 23,377,307 9.4% Cote d'Ivoire 19,839,750 16.0% 3,166,527 2.6% 32.5% 6,453,504 2.6% Equatorial Guinea 736,296 15.4% 113,337 0.1% 28.2% 207,708 0.1% Ethiopia 91,728,849 15.5% 14,225,062 11.9% 36.4% 33,348,005 13.4% Gabon 1,632,572 15.1% 246,008 0.2% 30.2% 492,829 0.2% Gambia 1,791,225 16.8% 300,419 0.3% 34.5% 617,771 0.2% Ghana 25,366,462 13.5% 3,434,583 2.9% 34.2% 8,679,224 3.5% Guinea 11,451,273 16.1% 1,841,084 1.5% 36.9% 4,225,239 1.7% Kenya 43,178,141 15.6% 6,720,436 5.6% 35.0% 15,107,354 6.1% Lesotho 2,051,545 11.0% 225,763 0.2% 31.2% 640,452 0.3% Malawi 15,906,483 16.9% 2,683,137 2.2% 38.0% 6,047,379 2.4% Mozambique 25,203,395 18.0% 4,540,898 3.8% 36.4% 9,162,223 3.7% Niger 17,157,042 21.3% 3,657,243 3.1% 39.2% 6,720,659 2.7% Nigeria 168,833,776 17.2% 28,966,558 24.2% 33.0% 55,630,915 22.4% Rwanda 11,457,801 16.0% 1,838,104 1.5% 35.1% 4,016,848 1.6% Senegal 13,726,021 16.8% 2,307,411 1.9% 33.4% 4,580,114 1.8% Sierra Leone 5,978,727 13.2% 790,351 0.7% 34.5% 2,062,677 0.8% Swaziland 1,230,985 13.9% 170,942 0.1% 35.8% 440,796 0.2% Tanzania 47,783,107 16.9% 8,061,564 6.7% 35.6% 17,026,239 6.9% Togo 6,642,928 15.1% 1,005,752 0.8% 34.7% 2,304,245 0.9% Uganda 36,345,860 18.9% 6,879,464 5.7% 38.6% 14,038,817 5.7% Zimbabwe 13,724,317 14.7% 2,018,254 1.7% 33.7% 4,631,843 1.9% Total 707,559,629 16.9% 119,687,523 100% 35.1% 248,241,673 100% Note: Total population per country in 2012 retrieved from World Bank Databank (Oct 2014). Child population as a share of total population per country based on authors’ calculations using most recent DHS and MICS surveys.

Page 35: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

35

ANNEX 3 – Deprivation headcount rate by indicator and age-group

Dimension Indicator Deprivation headcount in %

0 to 4 years 5 to 17 years

Nutrition 40.3% - Infant and young child feeding 56.6% - Wasting (weight for height) 8.6% -

Health 55.8% - DPT immunisation (1-4 years) 37.4% - Skilled birth attendance 49.1% -

Education17 35.2% Compulsory school attendance - 22.6% Primary school attainment - 51.0%

Information 26.3% Information devices - 26.3%

Water 51.8% 50.8% Drinking water source 40.4% 38.9% Distance to water source 24.6% 25.2%

Sanitation 67.1% 66.0% Toilet type 67.1% 66.0%

Housing 44.2% 43.7% Floor and roof material 35.3% 32.2% Overcrowding 16.3% 20.2%

Protection from Violence 63.2% 62.7% Domestic violence 63.2% 62.7%

17 Compulsory school attendance is calculated for children at official compulsory school age, which varies from country to country. Primary school attainment calculated for children who have reached the age of entering lower secondary school, up until the age of 17. The official starting and ending age of compulsory school, as well as the official duration of primary school and starting age of lower secondary school retrieved from: http://stats.uis.unesco.org/unesco/TableViewer/tableView.aspx?ReportId=163. One year of delay in schooling is allowed when calculating deprivation rates, allowing for delayed entry in schooling or one year of repetition. See background material in the MODA web-portal (http://www.unicef-irc.org/MODA/) for information on compulsory and primary school duration per country.

Page 36: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

36

ANNEX 4 – Contribution of multidimensional deprivation ratio (H) and adjusted multidimensional

deprivation ratio (M) by country to the total deprivation ratio (children deprived in 2-5

dimensions)

Contribution to the total

multidimensional deprivation ratio (H, K=2)

Contribution to the total adjusted multidimensional deprivation ratio

(M0 , K=2)

Age 0-4 Age 5-17 All children Age 0-4 Age 5-17 All children

Benin 1.2% 1.3% 1.3% 1.0% 1.1% 1.1%

Burkina Faso 2.3% 2.5% 2.4% 2.1% 2.4% 2.3%

Burundi 1.4% 1.4% 1.4% 1.2% 1.3% 1.3%

Cameroon 2.5% 2.2% 2.3% 2.4% 2.0% 2.2%

Central African Republic 0.8% 0.7% 0.7% 0.8% 0.7% 0.8%

Chad 2.6% 2.5% 2.6% 3.1% 3.0% 3.0%

Comoros 0.1% 0.1% 0.1% 0.0% 0.1% 0.1%

Congo, Republic of 0.5% 0.4% 0.4% 0.5% 0.3% 0.4%

Congo DR 11.9% 12.0% 12.0% 12.5% 13.8% 13.3%

Cote d'Ivoire 2.3% 1.9% 2.0% 2.0% 1.6% 1.7%

Equatorial Guinea 0.1% 0.1% 0.1% 0.1% 0.0% 0.1%

Ethiopia 15.0% 18.7% 17.3% 17.2% 21.5% 19.9%

Gabon 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%

Gambia 0.2% 0.1% 0.1% 0.1% 0.1% 0.1%

Ghana 1.7% 2.2% 2.0% 1.4% 1.8% 1.7%

Guinea 1.5% 1.5% 1.5% 1.4% 1.4% 1.4%

Kenya 5.2% 5.7% 5.5% 5.0% 5.3% 5.2%

Lesotho 0.2% 0.2% 0.2% 0.2% 0.2% 0.2%

Malawi 2.4% 3.0% 2.8% 2.1% 3.0% 2.7%

Mozambique 4.0% 4.2% 4.2% 4.2% 4.4% 4.3%

Niger 3.6% 3.5% 3.6% 4.0% 3.7% 3.8%

Nigeria 21.9% 16.3% 18.3% 20.5% 14.5% 16.7%

Rwanda 0.9% 1.0% 1.0% 0.6% 0.8% 0.7%

Senegal 1.3% 1.2% 1.2% 1.2% 1.0% 1.1%

Sierra Leone 0.6% 0.8% 0.7% 0.6% 0.7% 0.7%

Swaziland 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%

Tanzania 7.4% 7.9% 7.7% 7.7% 7.5% 7.6%

Togo 0.8% 0.9% 0.8% 0.7% 0.8% 0.8%

Uganda 6.1% 6.3% 6.2% 5.9% 5.7% 5.8%

Zimbabwe 1.3% 1.3% 1.3% 1.1% 1.2% 1.2%

TOTAL 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Page 37: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

37

ANNEX 5 – Multidimensional deprivation ratios for all cut-off points, by country and age-group

Multidimensional deprivation ratios: all children below the age of 18

Deprivation headcount rate (H) Average deprivation

intensity of deprived (A) Adjusted deprivation headcount rate (M)

No. of deprivations 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5

TOTAL 86.4% 67.1% 45.3% 23.5% 2.6 3.1 3.7 4.3 0.46 0.42 0.33 0.20

Benin 86.0% 57.6% 28.1% 9.0% 2.1 2.7 3.4 4.2 0.36 0.31 0.19 0.08

Burkina Faso 87.6% 67.0% 42.2% 18.3% 2.5 3.0 3.5 4.2 0.44 0.40 0.30 0.15

Burundi 90.1% 66.0% 35.5% 12.5% 2.3 2.8 3.4 4.2 0.41 0.36 0.24 0.10

Cameroon 74.1% 51.5% 31.0% 14.5% 2.4 3.0 3.6 4.3 0.35 0.30 0.22 0.12

Central African Republic 87.5% 72.2% 55.7% 32.5% 2.9 3.4 3.8 4.3 0.52 0.49 0.42 0.28

Chad 95.4% 88.3% 75.8% 50.7% 3.4 3.6 3.9 4.3 0.66 0.64 0.59 0.44

Comoros 83.6% 47.1% 17.0% 4.0% 1.8 2.5 3.3 4.1 0.30 0.23 0.11 0.03

Congo 77.1% 49.6% 24.3% 8.5% 2.1 2.7 3.4 4.2 0.32 0.27 0.17 0.07

Congo DR 93.8% 83.1% 66.9% 41.5% 3.2 3.5 3.8 4.3 0.60 0.58 0.51 0.36

Cote d'Ivoire 80.1% 52.0% 25.8% 8.4% 2.1 2.7 3.4 4.2 0.34 0.28 0.18 0.07

Equatorial Guinea 79.1% 47.7% 20.2% 3.8% 1.9 2.5 3.2 4.1 0.30 0.24 0.13 0.03

Ethiopia 97.0% 90.1% 76.3% 50.2% 3.4 3.6 3.9 4.3 0.66 0.64 0.59 0.43

Gabon 63.0% 29.6% 10.2% 2.3% 1.7 2.4 3.2 4.1 0.21 0.14 0.07 0.02

Gambia 65.3% 32.7% 13.1% 3.5% 1.8 2.5 3.3 4.1 0.23 0.17 0.09 0.03

Ghana 72.3% 41.3% 18.5% 5.2% 1.9 2.6 3.3 4.1 0.28 0.21 0.12 0.04

Guinea 83.2% 61.2% 37.7% 15.0% 2.4 2.9 3.5 4.2 0.40 0.36 0.26 0.13

Kenya 85.5% 62.8% 37.6% 16.5% 2.4 2.9 3.5 4.2 0.41 0.37 0.27 0.14

Lesotho 84.7% 58.3% 32.8% 12.8% 2.3 2.8 3.5 4.2 0.38 0.33 0.23 0.11

Malawi 94.9% 79.2% 51.3% 21.4% 2.6 3.0 3.5 4.2 0.50 0.47 0.36 0.18

Mozambique 87.7% 74.8% 55.4% 29.8% 2.9 3.2 3.7 4.3 0.51 0.49 0.41 0.25

Niger 94.0% 85.0% 67.1% 37.0% 3.1 3.3 3.7 4.2 0.58 0.57 0.50 0.31

Nigeria 79.6% 53.5% 29.6% 12.4% 2.2 2.8 3.5 4.2 0.36 0.30 0.21 0.11

Rwanda 77.0% 40.2% 14.3% 3.1% 1.8 2.4 3.2 4.1 0.27 0.20 0.09 0.03

Senegal 73.0% 44.2% 23.1% 8.7% 2.1 2.8 3.4 4.2 0.30 0.24 0.16 0.07

Sierra Leone 86.5% 62.7% 36.9% 15.9% 2.4 2.9 3.5 4.2 0.41 0.36 0.26 0.13

Swaziland 68.2% 33.6% 11.6% 3.3% 1.7 2.5 3.3 4.1 0.23 0.17 0.08 0.03

Tanzania 92.0% 75.9% 50.7% 25.0% 2.7 3.1 3.6 4.2 0.50 0.47 0.37 0.21

Togo 84.0% 62.3% 35.3% 13.8% 2.4 2.8 3.5 4.2 0.40 0.35 0.24 0.12

Uganda 92.0% 73.6% 44.5% 18.1% 2.5 2.9 3.5 4.2 0.46 0.43 0.31 0.15

Zimbabwe 70.8% 48.6% 28.7% 11.5% 2.3 2.9 3.5 4.1 0.32 0.28 0.20 0.10

Page 38: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

38

Multidimensional deprivation ratios: all children below the age of 5

Deprivation headcount rate (H)

Average deprivation intensity of deprived (A)

Adjusted deprivation headcount rate (M)

No. of deprivations 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5

TOTAL 91.5% 74.6% 53.7% 29.4% 2.8 3.2 3.7 4.3 0.51 0.48 0.40 0.25

Benin 89.7% 64.5% 34.0% 12.0% 2.3 2.7 3.4 4.2 0.41 0.36 0.23 0.10

Burkina Faso 90.1% 69.5% 44.7% 19.6% 2.5 3.0 3.5 4.2 0.46 0.42 0.32 0.17

Burundi 92.8% 70.1% 39.5% 13.1% 2.3 2.8 3.4 4.2 0.44 0.39 0.27 0.11

Cameroon 84.1% 62.6% 41.5% 20.4% 2.5 3.1 3.6 4.3 0.43 0.39 0.30 0.17

Central African Republic 92.4% 78.9% 61.8% 37.8% 3.1 3.4 3.8 4.3 0.57 0.54 0.47 0.33

Chad 97.6% 92.2% 82.2% 59.4% 3.6 3.8 4.0 4.4 0.71 0.69 0.65 0.52

Comoros 89.5% 56.7% 23.7% 5.9% 2.0 2.5 3.3 4.1 0.35 0.29 0.16 0.05

Congo 87.1% 62.1% 34.4% 13.3% 2.3 2.8 3.5 4.2 0.40 0.35 0.24 0.11

Congo DR 97.0% 87.0% 69.1% 39.6% 3.1 3.4 3.7 4.3 0.61 0.59 0.52 0.34

Cote d'Ivoire 88.0% 63.8% 37.5% 14.2% 2.4 2.9 3.5 4.2 0.41 0.37 0.26 0.12

Equatorial Guinea 90.1% 62.1% 35.4% 7.9% 2.2 2.7 3.2 4.1 0.39 0.34 0.23 0.06

Ethiopia 98.3% 94.3% 84.7% 58.8% 3.6 3.7 3.9 4.3 0.70 0.70 0.66 0.50

Gabon 76.3% 41.2% 17.4% 4.7% 1.8 2.5 3.3 4.1 0.28 0.21 0.11 0.04

Gambia 80.2% 48.8% 23.2% 7.4% 2.0 2.7 3.4 4.2 0.32 0.26 0.16 0.06

Ghana 75.7% 44.3% 21.8% 6.9% 2.0 2.7 3.4 4.2 0.30 0.24 0.15 0.06

Guinea 92.1% 72.9% 50.1% 23.8% 2.7 3.1 3.6 4.3 0.49 0.45 0.36 0.20

Kenya 89.0% 69.3% 45.8% 23.4% 2.6 3.1 3.6 4.2 0.47 0.43 0.33 0.20

Lesotho 87.8% 67.7% 43.9% 18.5% 2.5 3.0 3.5 4.2 0.44 0.40 0.31 0.15

Malawi 94.9% 79.4% 48.9% 18.7% 2.6 2.9 3.5 4.2 0.49 0.46 0.34 0.16

Mozambique 92.1% 79.4% 61.1% 35.1% 3.0 3.3 3.7 4.2 0.55 0.53 0.45 0.30

Niger 96.0% 88.7% 75.7% 48.6% 3.4 3.6 3.8 4.3 0.65 0.63 0.58 0.42

Nigeria 90.1% 67.6% 42.6% 20.1% 2.5 3.0 3.6 4.3 0.45 0.41 0.31 0.17

Rwanda 77.8% 41.7% 14.6% 2.9% 1.8 2.4 3.2 4.1 0.28 0.20 0.09 0.02

Senegal 80.1% 50.5% 30.5% 14.9% 2.2 3.0 3.6 4.2 0.36 0.30 0.22 0.13

Sierra Leone 92.9% 72.7% 45.7% 19.4% 2.5 3.0 3.5 4.2 0.47 0.43 0.32 0.16

Swaziland 71.4% 37.2% 13.9% 4.9% 1.8 2.5 3.4 4.1 0.26 0.19 0.09 0.04

Tanzania 94.5% 82.4% 63.2% 37.0% 3.0 3.3 3.8 4.3 0.58 0.55 0.48 0.32

Togo 88.0% 68.1% 42.5% 18.4% 2.5 2.9 3.5 4.2 0.44 0.40 0.30 0.16

Uganda 94.2% 79.1% 56.2% 26.7% 2.8 3.1 3.6 4.3 0.53 0.50 0.40 0.23

Zimbabwe 79.1% 55.3% 34.9% 15.3% 2.4 3.0 3.5 4.2 0.37 0.33 0.25 0.13

Page 39: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

39

Multidimensional deprivation ratios: all children of age 5-17

Deprivation headcount rate (H) Average deprivation

intensity of deprived (A) Adjusted deprivation headcount rate (M)

No. of deprivations 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5 1-5 2-5 3-5 4-5

TOTAL 83.9% 63.5% 41.2% 20.7% 2.6 3.1 3.6 4.3 0.43 0.39 0.30 0.18

Benin 84.3% 54.7% 25.5% 7.7% 2.1 2.6 3.3 4.2 0.35 0.29 0.17 0.06

Burkina Faso 86.4% 65.7% 41.0% 17.6% 2.5 3.0 3.5 4.2 0.43 0.39 0.29 0.15

Burundi 88.7% 63.9% 33.5% 12.2% 2.3 2.7 3.4 4.2 0.40 0.35 0.23 0.10

Cameroon 69.3% 46.0% 25.9% 11.6% 2.2 2.9 3.6 4.2 0.31 0.27 0.18 0.10

Central African Republic 84.6% 68.3% 52.1% 29.3% 2.9 3.3 3.7 4.3 0.49 0.45 0.39 0.25

Chad 94.3% 86.2% 72.4% 45.9% 3.3 3.5 3.8 4.3 0.63 0.61 0.56 0.40

Comoros 81.2% 43.0% 14.2% 3.2% 1.7 2.4 3.2 4.1 0.28 0.21 0.09 0.03

Congo 71.6% 42.7% 18.7% 5.8% 2.0 2.6 3.4 4.1 0.28 0.22 0.13 0.05

Congo DR 92.2% 81.1% 65.8% 42.5% 3.2 3.5 3.9 4.3 0.59 0.57 0.51 0.37

Cote d'Ivoire 76.2% 46.2% 20.1% 5.5% 2.0 2.6 3.3 4.1 0.30 0.24 0.13 0.05

Equatorial Guinea 73.1% 39.8% 11.9% 1.6% 1.7 2.3 3.1 4.0 0.25 0.19 0.08 0.01

Ethiopia 96.5% 88.2% 72.7% 46.5% 3.3 3.5 3.8 4.3 0.64 0.62 0.56 0.40

Gabon 56.4% 23.8% 6.6% 1.0% 1.6 2.3 3.2 4.1 0.18 0.11 0.04 0.01

Gambia 58.1% 24.9% 8.1% 1.7% 1.6 2.4 3.2 4.1 0.19 0.12 0.05 0.01

Ghana 70.9% 40.2% 17.2% 4.6% 1.9 2.6 3.3 4.1 0.27 0.21 0.11 0.04

Guinea 79.3% 56.0% 32.3% 11.2% 2.3 2.8 3.4 4.2 0.36 0.32 0.22 0.09

Kenya 83.9% 59.8% 33.9% 13.4% 2.3 2.8 3.5 4.2 0.39 0.34 0.24 0.11

Lesotho 83.5% 54.9% 28.8% 10.8% 2.2 2.8 3.4 4.2 0.36 0.30 0.20 0.09

Malawi 95.0% 79.1% 52.4% 22.5% 2.7 3.0 3.5 4.2 0.51 0.48 0.37 0.19

Mozambique 85.6% 72.5% 52.6% 27.2% 2.9 3.2 3.7 4.3 0.49 0.46 0.38 0.23

Niger 92.9% 83.0% 62.4% 30.7% 3.0 3.2 3.6 4.2 0.55 0.53 0.45 0.26

Nigeria 74.2% 46.2% 22.9% 8.4% 2.1 2.7 3.4 4.2 0.31 0.25 0.16 0.07

Rwanda 76.6% 39.5% 14.2% 3.2% 1.7 2.5 3.3 4.1 0.27 0.19 0.09 0.03

Senegal 69.5% 41.1% 19.3% 5.6% 2.0 2.6 3.3 4.1 0.27 0.22 0.13 0.05

Sierra Leone 84.0% 58.9% 33.6% 14.5% 2.3 2.9 3.5 4.2 0.39 0.34 0.24 0.12

Swaziland 66.9% 32.2% 10.8% 2.7% 1.7 2.4 3.3 4.1 0.23 0.16 0.07 0.02

Tanzania 90.8% 72.8% 44.8% 19.3% 2.5 2.9 3.5 4.2 0.46 0.43 0.31 0.16

Togo 82.3% 59.8% 32.1% 11.8% 2.3 2.8 3.4 4.2 0.38 0.33 0.22 0.10

Uganda 91.0% 70.9% 38.7% 13.9% 2.4 2.8 3.4 4.2 0.43 0.39 0.27 0.12

Zimbabwe 67.2% 45.7% 26.0% 9.8% 2.2 2.8 3.4 4.1 0.30 0.26 0.18 0.08

Page 40: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

40

ANNEX 6 - Estimates and predictions of multidimensional child poverty and monetary poverty

Country

GDP per capita, $

PPP, 2012 (WB)

Total population

Share of

urban pop.

Share of children as % of

the total pop.2

No. of children below 18

Multi-dimensional

child deprivation

ratio

No. of children

deprived in 2-5

dimensions

% of total pop. living

on less than $1.25PPP a

day

No. of poor children (less

than $1.25PPP a

day)3

Countries included in the CC-MODA analysis 1 Benin $1,716 10,050,702 42.7% 54.5% 5,474,142 57.6% 3,153,106 51.6% 2,825,205 2 Burkina Faso $1,555 16,460,141 27.3% 54.3% 8,940,923 67.0% 5,990,418 44.5% 3,975,134 3 Burundi $750 9,849,569 11.2% 53.7% 5,285,932 66.0% 3,488,715 81.3% 4,298,520 4 Cameroon $2,596 21,699,631 52.7% 50.8% 11,012,695 51.5% 5,671,538 27.6% 3,040,605 5 CAR $948 4,525,209 39.3% 53.0% 2,399,503 72.2% 1,732,441 62.8% 1,507,608 6 Chad $2,038 12,448,175 22.1% 57.7% 7,187,904 88.3% 6,346,919 36.5% 2,625,022 7 Comoros $1,519 717,503 28.0% 47.4% 340,248 47.1% 160,257 46.1% 156,888 8 Congo $5,730 4,337,051 64.1% 48.8% 2,115,607 49.6% 1,049,341 32.8% 694,342 9 Congo DR $697 65,705,093 41.0% 54.1% 35,544,504 83.1% 29,537,483 87.7% 31,179,639 10 Côte d'Ivoire $2,795 19,839,750 52.0% 48.5% 9,620,031 52.0% 5,002,416 35.0% 3,370,859 11 Eq. Guinea $35,908 736,296 39.5% 43.6% 321,045 47.7% 153,138 N/A N/A 12 Ethiopia $1,240 91,728,849 18.2% 51.9% 47,573,067 90.1% 42,863,333 36.8% 17,502,131 13 Gabon $18,347 1,632,572 86.4% 45.3% 738,836 29.6% 218,696 6.1% 44,995 14 The Gambia $1,604 1,791,225 57.7% 51.3% 918,191 32.7% 300,248 33.6% 308,787 15 Ghana $3,732 25,366,462 52.1% 47.8% 12,113,807 41.3% 5,003,002 28.6% 3,463,337 16 Guinea $1,237 11,451,273 35.7% 53.0% 6,066,323 61.2% 3,712,590 40.9% 2,479,306 17 Kenya $2,189 43,178,141 24.4% 50.6% 21,827,789 62.8% 13,707,852 43.4% 9,466,712 18 Lesotho $2,432 2,051,545 25.8% 42.2% 866,215 58.3% 505,003 56.2% 486,986 19 Malawi $753 15,906,483 15.8% 54.9% 8,730,517 79.2% 6,914,569 72.2% 6,299,941 20 Mozambique $985 25,203,395 31.4% 54.4% 13,703,121 74.8% 10,249,935 60.7% 8,319,165 21 Niger $899 17,157,042 18.0% 60.5% 10,377,902 85.0% 8,821,217 40.8% 4,235,222 22 Nigeria $5,535 168,833,776 45.2% 50.1% 84,597,473 53.5% 45,259,648 54.4% 45,995,646 23 Rwanda $1,406 11,457,801 25.9% 51.1% 5,854,952 40.2% 2,353,691 63.0% 3,689,791 24 Senegal $2,212 13,726,021 42.8% 50.2% 6,887,525 44.2% 3,044,286 34.1% 2,345,891 25 Sierra Leone $1,610 5,978,727 38.9% 47.7% 2,853,028 62.7% 1,788,849 56.6% 1,615,670 26 Swaziland $6,502 1,230,985 21.4% 49.7% 611,738 33.6% 205,544 39.3% 240,413 27 Tanzania $1,685 47,783,107 29.5% 52.5% 25,087,804 75.9% 19,041,643 43.5% 10,908,177 28 Togo $1,337 6,642,928 38.5% 49.8% 3,309,997 62.3% 2,062,128 52.5% 1,736,424 29 Uganda $1,357 36,345,860 15.1% 57.6% 20,918,282 73.6% 15,395,855 37.9% 7,930,121 30 Zimbabwe $1,696 13,724,317 32.8% 48.5% 6,650,097 48.6% 3,231,947 N/A N/A Countries for which deprivation rates are predicted1

31 Angola $5,539 20,820,525 41.7% 52.0% 10,826,620 32.6% 3,529,562 43.4% 4,695,505 32 Botswana $7,255 2,003,910 56.7% 52.0% 1,042,028 8.6% 89,814 13.4% 139,736 33 Cape Verde $3,554 494,401 63.4% 52.0% 257,087 34.0% 87,434 13.7% 35,272 34 Eritrea $504 6,130,922 21.4% 52.0% 3,188,064 71.3% 2,273,581 N/A N/A 35 Guinea-Bissau $494 1,663,558 46.9% 52.0% 865,046 62.3% 538,947 48.9% 423,007 36 Liberia $414 4,190,435 48.5% 52.0% 2,179,015 63.4% 1,381,254 83.8% 1,825,143 37 Madagascar $443 22,293,914 33.2% 52.0% 11,592,778 74.0% 8,576,047 87.7% 10,163,389 38 Mali $696 14,853,572 37.6% 52.0% 7,723,819 68.3% 5,274,167 50.6% 3,909,025 39 Mauritania $1,043 3,796,141 58.0% 52.0% 1,973,984 55.8% 1,100,607 23.4% 462,504 40 Namibia $5,931 2,259,393 43.7% 52.0% 1,174,879 22.5% 264,094 23.5% 276,566 41 São Tomé & Prin. $1,400 188,098 63.3% 52.0% 97,810 50.2% 49,071 43.5% 42,577 42 South Africa $7,314 52,274,945 63.3% 52.0% 27,182,838 22.2% 6,031,499 9.4% 2,560,623 43 South Sudan $974 10,837,527 18.2% 52.0% 5,635,486 70.4% 3,967,045 N/A N/A 44 Sudan $1,695 37,195,349 33.3% 52.0% 19,341,486 69.1% 13,369,963 19.8% 3,829,614 45 Zambia $1,463 14,075,099 39.6% 52.0% 7,319,015 61.6% 4,512,022 74.3% 5,439,492

Total 30 countries 367,929,196 246,965,808 As % of total 100.00% 67.1%

Total 28 countries (excl. ZWE and GNQ) 360,958,054 243,580,722 180,742,538 As % of total 100.00% 67.5% 50.1%

Total sub-Saharan Africa (SSA - 45 countries)4 468,329,152 298,010,915 As % of total 100.00% 63.6%

Total SSA (41 countries - excl. ZWE, GNQ, ERI, SSD) 452,534,460 288,385,203 214,544,993 As % of total 100.00% 63.7% 47.4% 1 Multidimensional child deprivation rates have been predicted for the remaining 15 countries of sub-Saharan Africa using coefficients retrieved from an OLS regression model estimating the relationship between multidimensional deprivation rates and the GDP per capita of 28 countries included in the analysis, controlled for the share of the urban population and the size of the population of each country. See Table 2 for regression results and the formula used for calculations. The countries that have been excluded from the regression as outliers: Equatorial Guinea and Gabon. 2 Children as a share of the total population calculated using DHS/MICS data for 30 countries. For the remaining 15 countries, child population assumed to be equal to 52% of the total population of each country, which is the average of child population of the 30 countries analysed. 3 This estimate is not adjusted for demographics; generally, child poverty rates in developing countries are higher than those of the total population. 4 Although part of sub-Saharan Africa according to the World Bank classification, the following three countries are excluded from this analysis: Somalia (no data on per capita GDP), Mauritius and Seychelles.

Page 41: CC-MODA Cross Country Multiple Overlapping Deprivation … · 1 CC-MODA – Cross Country Multiple Overlapping Deprivation Analysis: Analysing Child Poverty and Deprivation in sub-Saharan

41