OPHI - Human Development Reports
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
OPHI
Oxford Poverty & Human
Development Initiative
Empowered lives.
Resilient nations.
Charting pathways out of
multidimensional poverty:
Achieving the SDGsThe team that created this report includes Sabina Alkire, Pedro
Conceição, Cecilia Calderón, Jakob Dirksen, Maya Evans, Rolando
Gonzales, Jon Hall, Admir Jahic, Usha Kanagaratnam, Maarit Kivilo,
Milorad Kovacevic, Fanni Kovesdi, Corinne Mitchell, Ricardo Nogales,
Anna Ortubia, Mónica Pinilla-Roncancio, Natalie Quinn, Carolina
Rivera, Sophie Scharlin-Pettee and Nicolai Suppa. Peer reviewers
include Enrique Delamonica, Ivan Gonzalez de Alba, Gonzalo Hernandez
Licona, Frances Stewart and Bishwa Tiwari. The team would like to
thank the editors and layout artists at Communications Development
Incorporated—led by Bruce Ross-Larson, with Joe Brinley, Joe Caponio,
Christopher Trott and Elaine Wilson.
For a list of any errors and omissions found subsequent to printing, please visit http://hdr.undp.org and https://ophi.org.uk/multidimensional-poverty-index/.
Copyright @ 2020
By the United Nations Development Programme and Oxford Poverty and Human Development InitiativeGlobal Multidimensional Poverty Index 2020
Charting pathways out of multidimensional poverty: Achieving the SDGs
OPHI
Oxford Poverty & Human
Development Initiative
Empowered lives.
Resilient nations.Contents
Introduction 1 9 In 52 of the 71 countries with both multidimensional and monetary poverty
data, the incidence of multidimensional poverty fell faster in absolute terms 13
Part I. The global Multidimensional Poverty Index 2 10 Overlaying trends in the incidence of national and international monetary and
multidimensional poverty provides a fuller picture of a country’s poverty
Key findings 3
situation: Colombia, Pakistan and Sierra Leone 14
What is the global Multidimensional Poverty Index? 4 11 South Asia and Sub-Saharan Africa had the largest annualized absolute
The global Multidimensional Poverty Index in 2020 5 reductions in multidimensional poverty 15
Trends in multidimensional poverty: Progress and challenges 5 12 Forty-seven countries are on track to halve multidimensional poverty by 2030,
and eighteen are off track if observed trends continue 16
Projections of multidimensional poverty 12
13 Under a conservative scenario of the impact of COVID-19 on school attendance
and a moderate scenario of the impact on nutrition, simulations indicate that the
Part II. The Sustainable Development Goals and the global
increase in deprivations because of COVID-19 may set poverty levels back by 9.1
Multidimensional Poverty Index 18
years, with an additional 490 million people falling into multidimensional poverty 17
Key findings 19 14 Sustainable Development Goals that link to the global Multidimensional
The wide scope of interlinkages 20 Poverty Index 21
The MPI and immunization 22 15 Interlinked deprivations across 107 countries 21
16 There is a negative correlation between immunization coverage and the
The intersectionality of multidimensional poverty in education 25
incidence of multidimensional poverty 23
The MPI and the rural‑urban divide 27 17 The percentage of people who are multidimensionally poor and deprived in
The MPI, climate change and the environment 28 child mortality is highest in Nigeria, which also has one of the lowest
percentages of DTP3 coverage globally 24
The MPI, work and employment 31
18 In Democratic Republic of the Congo, Ethiopia, Nigeria and Pakistan the
percentage of people living with a child who did not receive the third dose of
Notes and references 35
the DPT-HepB-Hib vaccine is highest among multidimensionally poor people 24
Notes 36 19 Sub-Saharan African countries have the highest percentages of people who are
References 38 multidimensionally poor and deprived in years of schooling and school attendance 25
20 In Mali the mean years of schooling of adults older than 25 is higher for men
than for women across all poverty groups 26
STATISTICAL TABLES
21 In Haiti the differences in mean years of schooling between women and men
1 Multidimensional Poverty Index: developing countries 41
who live in rural and urban areas in different poverty groups are clear 26
2 Multidimensional Poverty Index: changes over time based on harmonized
22 Of the 1.3 billion people who are multidimensionally poor, 1.1 billion people
estimates 44
—84.2 percent—live in rural areas 28
23 The percentage of people who are multidimensionally poor and deprived in
BOXES each indicator is always higher in rural areas than in urban areas 29
1 Definitions for measuring changes in multidimensional poverty 6 24 In Kenya the impact of recent natural disasters is greater in provinces with
2 Reducing multidimensional poverty in Sierra Leone during the Ebola crisis 6 higher multidimensional poverty 30
3 The global Multidimensional Poverty Index and the Sustainable Development Goals 20 25 The percentages of people who are multidimensionally poor and deprived in
access to clean cooking fuel, access to clean drinking water and access to
improved sanitation are highest in Sub-Saharan Africa 30
FIGURES 26 The average contribution of environmental indicators to the Multidimensional
1 Structure of the global Multidimensional Poverty Index 4 Poverty Index differs significantly between rural and urban areas and across
2 Poorer countries with the highest initial Multidimensional Poverty Index values regions 31
and countries with low values tend to have slower absolute reduction rates 5 27 Child labour is more prevalent in countries with higher multidimensional poverty 32
3 In Madagascar multidimensional poverty declined most slowly among children, 28 Higher employment in the agricultural sector is associated with higher
even though they were the poorest age group 7 multidimensional poverty in Sub-Saharan Africa 32
4 Some of the poorest countries in Sub-Saharan Africa achieved the fastest 29 A higher share of informal employment in nonagricultural employment is
absolute reductions in multidimensional poverty 8 associated with higher multidimensional poverty 33
5 Reductions in multidimensional poverty can be driven by improvements in 30 In countries with high multidimensional poverty a large share of the
different indicators 8 population lacks any social protection 34
6 Absolute and relative annualized reductions in multidimensional poverty 10
7 Bangladesh, Lao People’s Democratic Republic and Mauritania show a TABLE
pro-poor trend in reducing multidimensional poverty 11
1 COVID-19 scenarios, projected global Multidimensional Poverty Index values,
8 Which country reduced each indicator fastest and when? 12 increases in the number of multidimensionally poor people, and length of setback 17
ii | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020Introduction
The lives of poor people are an intricate balance; their steps out of poverty even more so. Millions of daily labourers, herders
and farmers eking out subsistence on rugged terrain have no access to clean drinking water and no electric light at home.
Street vendors’ children may be undernourished, and entire families illiterate. In tough times many children drop out of
school. Improvements may come—an electrification scheme, better water and sanitation, upgraded schools with lunch
programmes, and good local health clinics. But conflicts, migrations, disasters and shocks also threaten.
Launched in 2010 by the Oxford Poverty and greater environmental threats. By detailing
and Human Development Initiative at the the connections between the MPI and other
University of Oxford and the Human Devel- poverty-related SDGs, the report highlights
opment Report Office of the United Nations how the lives of multidimensionally poor peo-
Development Programme for the flagship ple are precarious in ways that extend beyond
Human Development Reports, the global the MPI’s 10 component indicators.
Multidimensional Poverty Index (MPI) meas- The COVID-19 pandemic unfolded in the
ures the complexities of poor people’s lives, midst of this analysis. While data are not yet
individually and collectively, each year. This available to measure the rise of global poverty
report—released 10 years after that launch after the pandemic, simulations based on dif-
—focuses on how multidimensional poverty ferent scenarios suggest that, if unaddressed,
has declined. It provides a comprehensive progress across 70 developing countries could
picture of global trends in multidimensional be set back 3–10 years.
poverty, covering 5 billion people. It probes The firm hope is that it will not. As Amart-
patterns between and within countries and by ya Sen observes, Britain during World War II
indicator, showcasing different ways of making suffered food shortages and an overall decline
progress. Together with data on the $1.90 a in food availability. Yet with judicious rationing
day poverty rate, the trends monitor global and proactive policies, life expectancy rose. In
poverty in different forms. the decade before the war, life expectancy had
This is a key moment to study how nonmon- risen by 1.2 years for men and by 1.5 years for
etary poverty goes down. It is 10 years before women. But during the war it rose by 6.5 years
2030, the due date of the Sustainable Develop- for men and by 7 years for women.1 Evidence
ment Goals (SDGs), whose first goal is to end suggests a similar story in Sierra Leone, which
poverty in all its forms everywhere. And it is a had the fastest reduction in MPI value among
year when a pandemic and economic slowdown all countries with trend data. And this occurred
are pushing many more into poverty, while the during the Ebola crisis, not after. One by one
spectre of racism still haunts, and environmen- these stories seem tenuous, even improbable.
tal threats such as locusts surge. But the hope is that the information on mul-
Multidimensional poverty is strongly associ- tidimensional poverty summarized here and
ated with other SDG challenges. Concentrated detailed online will encourage and empower
in rural areas, multidimensionally poor people readers to fight to end poverty during these
tend to experience lower vaccination rates and difficult times, even against all odds. If they do,
secondary school achievement, insecure work progress is possible.
Charting pathways out of multidimensional poverty: Achieving the SDGs | 1PART I
The global Multidimensional
Poverty IndexKey findings reduction in the number of multidimension-
ally poor people (273 million). Ten coun-
• Across 107 developing countries, 1.3 billion tries, including China, came close to halving
people—22 percent—live in multidimen- their MPI value.3
sional poverty.2 • In nearly a third of the countries studied,
• Children show higher rates of multidimen- either there was no reduction in multidimen-
sional poverty: half of multidimensionally sional poverty for children, or the MPI value
poor people (644 million) are children under fell more slowly for children than for adults.
age 18. One in three children is poor com- • The countries with the fastest reduction
pared with one in six adults. in MPI value in absolute terms were Sierra
• About 84.3 percent of multidimensionally Leone, Mauritania and Liberia, followed by
poor people live in Sub-Saharan Africa Timor-Leste, Guinea and Rwanda. North
(558 million) and South Asia (530 million). Macedonia had the fastest relative poverty
• 67 percent of multidimensionally poor reduction, followed by China, Armenia,
people are in middle-income countries, Kazakhstan, Indonesia, Turkmenistan and
where the incidence of multidimensional Mongolia. Each of these countries cut its
poverty ranges from 0 percent to 57 percent original MPI value by at least 12 percent a
nationally and from 0 percent to 91 percent year.
subnationally. • In 14 countries in Sub-Saharan Africa, the
• Every multidimensionally poor person is be- number of multidimensionally poor people
ing left behind in a critical mass of indicators. increased, even though their MPI value de-
For example, 803 million multidimensionally creased, because of population growth.
poor people live in a household where some- • How countries reduced their MPI value var-
one is undernourished, 476 million have an ies by indicator and by subnational region.
out-of-school child at home, 1.2 billion lack Twenty countries significantly reduced dep-
access to clean cooking fuel, 687 million lack rivations for every indicator. Bangladesh, Lao
electricity and 1.03 billion have substandard People’s Democratic Republic and Maurita-
housing materials. nia had pro-poor reductions in subnational
• 107 million multidimensionally poor regions.
people are age 60 or older—a particularly • Multidimensional poverty trends do not
importantly figure during the COVID-19 match monetary poverty trends, suggesting
pandemic. different drivers.
• 65 countries reduced their global Multidi- • Charting trends in multidimensional and
mensional Poverty Index (MPI) value signif- monetary poverty measures and using glob-
icantly in absolute terms. Those countries are al data and national statistics, as Atkinson
home to 96 percent of the population of the (2019) proposed, provides an overall picture
75 countries studied for poverty trends. The of a country’s poverty situation.
fastest, Sierra Leone (2013–2017), did so • Before the pandemic 47 countries were on
during the Ebola epidemic. track to halve poverty between 2015 and
• Four countries halved their MPI value. India 2030, if observed trends continued. But 18
(2005/2006–2015/2016) did so nationally countries, including some of the poorest,
and among children and had the biggest were off track.
Charting pathways out of multidimensional poverty: Achieving the SDGs | 3The 2020 global Multidimensional Poverty terms, poverty is now understood to include
Index (MPI) provides current levels of multi- the lived reality of people’s experiences and the
dimensional poverty in developing countries, multiple deprivations they face. Since 2010 the
the highlights of which are listed in the key global MPI has compared acute multidimen-
findings. Part I first introduces the global sional poverty across more than 100 countries.
MPI and presents trends in poverty reduction The global MPI examines each person’s dep-
for 5 billion people living in a subset of those rivations across 10 indicators in three equally
countries. It then presents projections to an- weighted dimensions—h ealth, education
swer two pressing questions: Are countries on and standard of living (figure 1) and offers a
track to halve poverty by 2030, and how might high-resolution lens to identify both who is
their poverty be affected by the COVID-19 poor and how they are poor. It complements
pandemic? the international $1.90 a day poverty rate by
showing the nature and extent of overlapping
deprivations for each person.
What is the global In the global MPI, people are counted as
Multidimensional Poverty Index? multidimensionally poor if they are deprived
in one-third or more of 10 indicators (see fig-
Sustainable Development Goal (SDG) 1 aims ure 1), where each indicator is equally weighted
to end poverty in all its forms everywhere.4 within its dimension, so the health and educa-
Although previously defined only in monetary tion indicators are weighted 1/6 each and the
FIGURE 1
Structure of the global Multidimensional Poverty Index
Nutrition
Health
Child mortality
Years of schooling
Three
dimensions Education
of poverty
School attendance
Cooking fuel
Sanitation
Standard Drinking water
of living Electricity
Housing
Assets
Source: OPHI 2018.
4 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020standard of living indicators are weighted 1/18 global MPI, home to roughly 5 billion people
each. The intensity of multidimensionally poor across all developing regions (figure 2).11 The
people is measured by the average number of timespan for the analysis ranges from 3 years
weighted deprivations they experience. The between surveys to 12 years. The MPI estimates
MPI is the product of the incidence of poverty used in this section are rigorously harmonized
(proportion of poor people) and the intensity and denoted by MPIT so indicator definitions
of poverty (average deprivation score5 of poor match between time periods (for example, if
people) and is therefore sensitive to changes in one survey collected only child nutrition rather
both components. The MPI ranges from 0 to 1, than adult nutrition, data for the other survey
and higher values imply higher poverty. are restricted to child nutrition as well).12 Due
To ensure transparency, the detailed defini- to this harmonization, the MPIT values in sta-
tion of each indicator is published online, with tistical table 2 may differ from those in statis-
country-specific adjustments and the computer tical table 1 (which represents the best possible
code used to calculate the global MPI value for MPI estimate that can be calculated with the
each country.6 information available).13 Box 1 defines key
terms used in the discussion of poverty trends.
The global Multidimensional FIGURE 2
Poverty Index in 2020
Poorer countries with the highest initial
Multidimensional Poverty Index values and
The 2020 update of the global Multidimensional countries with low values tend to have slower
Poverty Index (MPI) covers 107 countries—28 absolute reduction rates
low income, 76 middle income and 3 high
income7—and 5.9 billion people in developing Arab States
East Asia and the Pacific
regions. MPI values and data for the MPI’s com- Europe and Central Asia
ponent indicators are also disaggregated by age Latin America and the Caribbean
group, for rural and urban areas and for 1,279 South Asia
Sub-Saharan Africa
subnational regions. Data for 25 countries cover-
ing 913 million people have been updated from MPIT value
0.700
the 2019 release.8 The 2020 estimates are based
on 47 Demographic and Health Surveys (DHS),
0.600
47 Multiple Indicator Cluster Surveys (MICS),
3 Pan Arab Population and Family Health Ethiopia
Surveys and 10 national surveys. All surveys are 0.500
dated 2008–2019, and data for 83 countries— Sierra Leone
home to 92 percent of multidimensionally poor 0.400
people—were collected in 2013/2014 or later.9 Nigeria
The global MPI is thus a key resource for recent 0.300
India
poverty data across developing regions.
0.200
Trends in multidimensional 0.100
China
poverty: Progress and challenges
0.000
2000 2005 2010 2015 2020
While the United Nations Development Pro- Year of the survey
gramme has previously published tables with
estimates on global trends in multidimensional MPIT is the Multidimensional Poverty Index estimate that is based on harmonized
poverty, this is the first study that focuses on indicator definitions for strict comparability over time.
Note: The figure shows the level of multidimensional poverty in the starting and
harmonized trends to shed light on the dynam- ending periods of the study. The size of each bubble represents the number of
multidimensionally poor people in each year, the colour indicates the region of the
ics of poverty reduction and to increase un- country and the trend line connecting the bubbles depicts the speed of reduction.
derstanding of what is possible.10 The analysis The horizontal placement refers to the years of the surveys.
Source: Alkire, Kovesdi, Mitchell and others 2020.
covers 75 of the 107 countries included in the
Charting pathways out of multidimensional poverty: Achieving the SDGs | 5BOX 1
Definitions for measuring changes in multidimensional poverty
Absolute change (annualized). The difference in a pov- Headcount ratio (also called incidence) is the most fa-
erty measure between two years, divided by the number miliar measure. It shows the change in the percentage
of years between surveys. of people who are multidimensionally poor (but not the
intensity of poverty or the number of poor people).
Relative change (annualized). The compound rate of
change per year.1 It shows the percentage by which the Intensity shows how the average deprivation score of
previous year’s poverty has changed. poor people has changed.
What has changed? Changes in… Number of multidimensionally poor people (calcu-
lated as the product of the incidence of multidimensional
Multidimensional Poverty Index value (MPIT value) poverty and the population size) shows how the overall
is the most comprehensive measure of multidimension- number of multidimensionally poor people in a country
al poverty. It considers changes in both the incidence has changed and reflects both demographic change and
and the intensity of poverty (but not the number of poor population growth (but not the MPIT or the intensity of
people). poverty). It is important for budgeting and targeting.
Note
1. The compound rate of change is the geometric progression ratio that provides a constant rate of return over the time period.
Sixty-five countries significantly countries in the middle of the distribution, with
reduced multidimensional poverty moderate to high MPI values had the fastest
reductions. Overall, 62 of the 65 countries with
Sixty-five countries, home to 96 percent of the a significant reduction in MPI value had a sig-
population of the 75 countries studied, signifi- nificant reduction in the incidence of poverty.
cantly reduced multidimensional poverty.14 The In 23 of those countries, more than 2 percent of
fastest country, Sierra Leone (2013–2017) did the population moved out of poverty every year
so during the Ebola epidemic (box 2). during the included period—rising to nearly
Figure 2 depicts changes in MPIT value for 4 percent a year in Sierra Leone. The incidence
all 75 countries between two periods of time. of poverty in these countries in their starting
Poorer countries with the highest initial MPI year ranged from 20 percent in Mongolia to
values and countries with low MPI values tend 82 percent in Liberia, showing that progress
to have slower absolute reduction rates, whereas is possible across countries with very different
BOX 2
Reducing multidimensional poverty in Sierra Leone during the Ebola crisis
From December 2013 to March 2016, the Ebola crisis largest annualized absolute reduction in depriviation in
spread in West Africa. As terrible as the tragedy was, it clean cooking fuel and in child mortality among the 75
did not create a widespread slide into poverty. The fast- countries studied. It had the fastest absolute reduction
est reduction in multidimensional poverty among the in MPI value among children of all countries, though
75 countries studied, covering nearly 5 billion people, poverty among adults declined faster. And although the
was in Sierra Leone, where the percentage of people in poorest regions did not move the fastest, 12 of Sierra
multidimensional poverty fell from 74 percent in 2013 Leone’s 14 subnational regions reduced their MPIT value.
to 58 percent in 2017—the same years as the Ebola Public health emergencies require fast responses,
crisis. The percentage of people who were multidimen- and human error as well as tragedy seem inevitable.
sionally poor and deprived declined for all 10 indicators, Despite this, Sierra Leone shows that it is possible to
with the biggest reductions related to deprivations in reduce the interlinked deprivations of multidimensional
cooking fuel and electricity. Sierra Leone also had the poverty during an epidemic.
6 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020poverty rates. The remaining countries moved FIGURE 3
more slowly.
In Madagascar multidimensional poverty declined
most slowly among children, even though they
Halving multidimensional were the poorest age group
poverty is possible
MPIT value, 2008/09
Four countries—Armenia (2010–2015/2016), 0.100 0.200 0.300 0.400 0.500 0.600
India (2005/2006–2015/2016), Nicaragua 0.000
(2001–2011/2012) and North Macedonia
–0.002
(2005/2006–2011) halved their global MPIT
value and did so in 5.5–10.5 years. These coun- Ages 0–17
–0.004
tries show what is possible for countries with
very different initial poverty levels. They account Ages 65 and older
–0.006
for roughly a fifth of the world’s population,
mostly because of India’s large population.15 Ten –0.008 Ages 18–64
countries—including China and Indonesia— Annualized
came close to halving their level of multidimen- absolute change
sional poverty (MPIT).16 Only two countries MPIT is the Multidimensional Poverty Index estimate that is based on harmonized
(Nicaragua and North Macedonia) halved the indicator definitions for strict comparability over time.
Source: Alkire, Kovesdi, Mitchell and others 2020.
incidence of multidimensional poverty. SDG
1 and the Third Decade on Poverty Reduction
call for ending multidimensional poverty using 10 and 10.5 years respectively, and during that
integrated approaches and policy frameworks;17 time both countries halved their MPIT values
these trends show that progress is possible. among children. So decisive change for children
is possible but requires conscious policy efforts.
Some reductions overlook children
Some of the poorest countries
Across the 75 countries studied, nearly half of in Sub-Saharan Africa achieved
poor people are children under age 18. But in the fastest absolute reductions
nearly a third of these countries, either there in multidimensional poverty
was no reduction in multidimensional poverty
for children (ages 0–17), or the MPI value fell Sub-Saharan African countries have the highest
more slowly for children than for adults (ages poverty rates and some of the bleakest prog-
18–64). In 13 countries there was no statisti- noses. Several of these countries struggle with
cally significant reduction in multidimensional political conflicts, violence, environmental
poverty among children.18 And in 11 of the problems and rapid population growth. Yet
60 countries with a significant reduction for some of the poorest countries in Sub-Saharan
both age groups—all of them in Sub-Saharan Africa are among those with the fastest abso-
Africa—the reduction in poverty was faster for lute reduction in multidimensional poverty
adults than for children. This includes Mad- (figure 4).
agascar, where multidimensional poverty fell Sierra Leone, Mauritania and Liberia re-
most slowly among children, even though they duced their MPI value fastest. Mauritania start-
were the poorest age group (figure 3). A focus ed with a multidimensional poverty headcount
on children is critical, as in 13 of the 60 coun- of 63 percent, Sierra Leone with a headcount
tries studied there was no reduction in child of 74.1 percent and Liberia with a headcount
poverty, and these countries span every major of 81.6 percent. Their success was driven by
geographic region except South Asia as well as improvements in different indicators (figure 5).
low to high levels of MPI. In Sierra Leone (2013–2017), deprivations in
On a positive note, Mauritania, Sierra Leo- nutrition, school attendance, cooking fuel, san-
ne, Timor-Leste, Liberia and Rwanda had the itation, water, electricity and housing all fell by
fastest reduction in child poverty in absolute more than 2 percentage points a year. In Mau-
terms. India and Nicaragua’s time periods cover ritania (2011–2015), improvement in years of
Charting pathways out of multidimensional poverty: Achieving the SDGs | 7FIGURE 4
Some of the poorest countries in Sub-Saharan Africa achieved the fastest absolute reductions in multidimensional poverty
Congo, Democratic Republic of the (2007–2013/2014)
Tanzania (United Republic of) (2010–2015/2016)
Sao Tome and Principe (2008/2009–2014)
Central African Republic (2000–2010)
Eswatini (Kingdom of) (2010–2014)
Madagascar (2008/2009–2018)
Côte d’Ivoire (2011/2012–2016)
Zimbabwe (2010/2011–2015)
Rwanda (2010–2014/2015)
Namibia (2006/2007–2013)
Gambia (2005/2006–2013)
Burundi (2010–2016/2017)
Malawi (2010–2015/2016)
Zambia (2007–2013/2014)
Burkina Faso (2006–2010)
Congo (2005–2014/2015)
Kenya (2008/2009–2014)
Mozambique (2003–2011)
Sierra Leone (2013–2017)
Chad (2010–2014/2015)
Mauritania (2011–2015)
Senegal (2005–2017)
Lesotho (2009–2014)
Ethiopia (2011–2016)
Uganda (2011–2016)
Nigeria (2013–2018)
Guinea (2012–2016)
Liberia (2007–2013)
Gabon (2000–2012)
Ghana (2011–2014)
Niger (2006–2012)
Mali (2006–2015)
0.000
–0.005
–0.010
–0.015
–0.020
–0.025
–0.030
Annualized absolute change in MPIT value
MPIT is the Multidimensional Poverty Index estimate that is based on harmonized indicator definitions for strict comparability over time.
Source: Alkire, Kovesdi, Mitchell and others 2020.
FIGURE 5
Reductions in multidimensional poverty can be driven by improvements in different indicators
Sierra Leone (2013–2017) Mauritania (2011–2015) Liberia (2007–2013)
Child Years of School Cooking Drinking
Nutrition mortality schooling attendance fuel Sanitation water Electricity Housing Assets
0
–1
–2
–3
–4
–5
–6
Annualized absolute change in percentage of people who are multidimensionally poor and deprived in each indicator (percentage points)
Source: Alkire, Kovesdi, Mitchell and others 2020.
schooling was the main factor. Deprivations in school attendance and asset ownership drove
in school attendance, sanitation, and drinking the reduction. Deprivations in cooking fuel,
water also fell by more than 2 percentage points sanitation and electricity also fell by more than
a year. In Liberia (2007–2013), improvements 2 percentage points a year.
8 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020Strong reductions in multidimensional number of poor people rose because of rapid
poverty in East Asia and Pacific population growth. In Niger, the country with
the highest MPI value, the population grew
East Asia and Pacific and Europe and Central by a quarter in six years, and the number of
Asia boast notable examples of MPIT reduc- multidimensionally poor people increased by
tion relative to starting levels (figure 6). China 21.7 percent, despite reductions in both the
(2010–2014) led East Asia and Pacific, with an incidence and the intensity of multidimension-
annual relative reduction of over 19 percent, al poverty. These findings show the impact of
lifting more than 70 million people out of population growth on the number of multidi-
poverty in just four years, thanks to substantial mensionally poor people.
improvements in nutrition, access to drinking
water, clean cooking fuel, education and asset Leaving no one behind: When the
ownership. Indonesia (2012–2017), anoth- poorest subnational regions reduce
er populous country, reduced incidence by multidimensional poverty the fastest
12.2 percent a year, and 17 of its 33 subnational
regions halved their MPIT value in merely The SDGs aim to make equitable progress—
five years. In relative terms, Thailand and Lao which means prioritizing interventions for the
People’s Democratic Republic reduced their poorest of the poor. Of the 625 subnational
MPIT value by about 10 percent a year, and regions included in the analysis, 398—home
Indonesia, Lao People’s Democratic Republic to over three-quarters of multidimensionally
and Timor-Leste had statistically significant poor people in both periods—had statistical-
decreases in the percentage of people who were ly significant decreases in their MPIT value.
multidimensionally poor and deprived in every Fourteen countries reduced multidimensional
indicator. Only seven years after receiving for- poverty in all their subnational regions: Bang-
mal UN recognition, Timor-Leste reduced the ladesh, Bolivia, the Kingdom of Eswatini,
incidence of multidimensional poverty from Gabon, Gambia, Guyana, India, Liberia, Mali,
69.6 percent in 2009/2010 to 46.9 percent in Mozambique, Niger, Nicaragua, Nepal and
2016, the fastest absolute reduction in East Rwanda.
Asia and Pacific and the fourth fastest among Disaggregating the global MPIT by subna-
the 75 countries studied. tional region shows whether the poorest areas
are making faster progress. Bangladesh and Lao
Fewer multidimensionally poor People’s Democratic Republic show a clear pro-
people in many countries— poor trend, with the poorest regions generally
but not in all countries reducing their MPIT value the fastest in abso-
lute terms (figure 7). Still, the poorest region of
Of the 65 countries that reduced their MPIT Lao People’s Democratic Republic (Saravane)
value, 50 also reduced the number of people —which had more poor people than the three
living in poverty. The largest reduction was in next-poorest regions—did not have the fastest
India, where approximately 273 million people progress.
moved out of multidimensional poverty over
10 years.19 In China more than 70 million Every indicator makes a difference
people moved out of multidimensional poverty
over four years, and 19 million people in Bang- All 10 of the indicators on which the MPI is
ladesh and almost 8 million people in Indo- based played a role in reducing poverty. Of
nesia did so over five years. In Pakistan almost the 75 countries studied, 20 significantly re-
4 million people moved out of poverty over duced deprivations in every indicator, and 11
five years. Some smaller countries also achieved of those were in Sub-Saharan Africa.20 Of the
a remarkable reduction: almost 4 million in 625 subnational regions studied, 45 reduced
Nepal and more than 3 million in Kenya over deprivations in every indicator. Figure 8 dis-
five years. plays the countries that had the largest absolute
Nevertheless, in 14 Sub-Saharan African reduction in deprivation for each of the 10
countries that reduced their MPIT value, the indicators. All these countries are low income
Charting pathways out of multidimensional poverty: Achieving the SDGs | 9FIGURE 6
Absolute and relative annualized reductions in multidimensional poverty
Absolute annualized change in MPIT value Relative annualized reduction in MPIT value (%)
–0.030 –0.025 –0.020 –0.015 –0.010 –0.005 0.000 0 5 10 15 20 25
Sierra Leone North Macedonia
Mauritania China
Liberia Armenia‡
Timor-Leste Kazakhstan
Guinea Indonesia
Rwanda Turkmenistan
Lao People’s Democratic Republic Mongolia
Sao Tome and Principe Sao Tome and Principe
Côte d’Ivoire Lao People’s Democratic Republic
Honduras Kyrgyzstan
India Honduras
Malawi Thailand‡
Nepal Suriname
Congo Eswatini (Kingdom of)
Bangladesh Albania
Mozambique Bolivia (Plurinational State of)
Cambodia Bosnia and Herzegovina
Bolivia (Plurinational State of) Bangladesh
Lesotho Dominican Republic
Gambia Tajikistan
Nicaragua Guyana
Uganda Nicaragua
Niger Peru
Kenya Nepal
Zambia Egypt
Eswatini (Kingdom of) Belize
Ethiopia Congo
Ghana Ghana
Tanzania (United Republic of) India
Haiti Timor-Leste
Mali Mauritania
Sudan Sierra Leone
Central African Republic Moldova (Republic of)
Mongolia Lesotho
Burundi Cambodia
Senegal Philippines
Burkina Faso Rwanda
Congo, Democratic Republic of the Iraq
Namibia Gabon
Pakistan Côte d’Ivoire
Nigeria Kenya
Madagascar Guinea
Zimbabwe Liberia
Gabon Malawi
China Mexico
Suriname Haiti
Chad Uganda
North Macedonia Gambia
Tajikistan Zimbabwe
Peru Zambia
Iraq Namibia
Indonesia Colombia
Kyrgyzstan Tanzania (United Republic of)
Dominican Republic Pakistan
Philippines Mozambique
Belize Sudan
Egypt Senegal
Guyana Nigeria
Mexico Ethiopia
Bosnia and Herzegovina Mali
Turkmenistan Niger
Colombia Burundi
Albania Congo, Democratic Republic of the
Thailand‡ Central African Republic
Kazakhstan Madagascar
Moldova (Republic of) Burkina Faso
Armenia‡ Chad
‡
indicates that the change in MPIT is significant only at 90 percent.
PIT is the Multidimensional Poverty Index estimate that is based on harmonized indicator definitions for strict comparability over time.
M
Source: Alkire, Kovesdi, Mitchell and others 2020.
10 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020FIGURE 7 except Timor-Leste and Mauritania, which are
lower middle income.
Bangladesh, Lao People’s Democratic Republic
The starting and ending percentages for child
and Mauritania show a pro-poor trend in reducing
multidimensional poverty mortality are by far the lowest because this tragic
deprivation has the lowest incidence, so its re-
Lao People’s Democratic Republic duction is the smallest. The proportional reduc-
2011/2012–2017 tion of deprivations is smallest in cooking fuel.
MPIT value, 2011/2012
0.000 0.100 0.200 0.300 0.400 Trends in multidimensional and
0.000 Vientiane Capital monetary poverty— different
–0.010
but complementary
–0.020 Trends in multidimensional poverty com-
Saravane plement trends in monetary ($1.90 a day)
–0.030 poverty.21 In 52 of the 71 countries with both
Phongsaly multidimensional and monetary poverty data,
–0.040 the incidence of multidimensional poverty fell
Annualized
absolute change faster in absolute terms (figure 9), while the
incidence of monetary poverty fell faster in 19
Bangladesh 2014–2019 countries. The difference was particularly strik-
MPIT value, 2014
0.100 0.150 0.200 0.250 0.300
ing in the Arab States, where every country saw
–0.010 Dhaka
either a slower reduction in monetary poverty
than in multidimensional poverty or an increase
–0.015 Khuina in monetary poverty. Some of the poorest
countries, such as Niger and Chad, saw larger
–0.020 reductions in monetary poverty than in multi-
dimensional poverty. This is partly because the
–0.025 Sylhet
incidence of multidimensional poverty (which
can be compared with the monetary poverty
–0.030
Annualized headcount ratio) does not tell the entire story.
absolute change Niger had the fourth fastest absolute reduction
in the intensity of multidimensional poverty
Mauritania 2011–2015 and reduced deprivation across all 10 indicators
MPIT value, 2011
—a reduction captured by the MPIT but not by
0.000 0.100 0.200 0.300 0.400 0.500 0.600
0.000
the trends in the MPIT headcount ratio.
Tagant
Nouakchott Triangulating poverty trends to
–0.020 reveal the overall picture
–0.040 Gorgol Overlaying trends in national and international
monetary poverty measures and national and
Hosh el Gharbi international multidimensional poverty meas-
–0.060
Annualized ures in one place—as suggested by the late Sir
absolute change
Tony Atkinson, a leading voice in poverty and
MPIT is the Multidimensional Poverty Index estimate that is based on harmonized
inequality measurement, in Measuring Poverty
indicator definitions for strict comparability over time. around the World22—can provide a fuller pic-
Note: Regions are ordered horizontally from the least poor in terms of their
starting MPIT value on the left to the poorest on the right, and vertically from the ture of a country’s poverty situation. Figure 10
slowest absolute progress on the top to the fastest at the bottom. The size of
the bubbles indicates the number of multidimensionally poor people in the initial
presents this analysis for three countries—
period. Grey bubbles indicate that no statistically significant change in MPIT Colombia, Pakistan and Sierra Leone—in
occurred for that region.
Source: Alkire, Kovesdi, Mitchell and others 2020. different regions and with different incidences
of poverty.23 In Colombia multidimensional
poverty measured by national definitions fell
Charting pathways out of multidimensional poverty: Achieving the SDGs | 11FIGURE 8
Which country reduced each indicator fastest and when?
Share of people who are multidimensionally poor Year 1 Year 2
and deprived in each indicator (%)
80
70
60
50
40
30
20
10
0
Nutrition
Rwanda (2010–2014/2015)
Child mortality
Sierra Leone (2013–2017)
Years of schooling
Mauritania (2011–2015)
School attendance
Liberia (2007–2013)
Cooking fuel
Sierra Leone (2013–2017)
Sanitation
Malawi (2011–2015/2016)
Drinking water
Timor-Leste (2009/2010–2016)
Electricity
Timor-Leste (2009/2010–2016)
Housing
Guinea (2012–2016)
Assets
Liberia (2007–2013)
Note: The height of the bar indicates the percentage of people who were multidimensionally poor and deprived in that indicator at the beginning of the survey period, and
the orange portion of the bar indicates the percentage by the end of the survey period.
Source: Alkire, Kovesdi, Mitchell and others 2020.
faster than monetary poverty. The incidence Projections of
of multidimensional poverty according to the multidimensional poverty
global MPIT in Colombia is low, suggesting
the need for a global measure of moderate The estimates of changes in multidimension-
poverty in addition to the existing measure of al poverty over time can be used to project
acute poverty (global MPI). In Pakistan trends whether countries are on track to achieve the
in the incidence of multidimensional poverty SDG target of at least halving the proportion
according to national definitions and the global of people living in poverty in all its dimensions
MPIT suggest that multidimensional poverty by 2030 as well as the possible impacts of
fell more slowly than monetary poverty. Trend COVID-19.24
data are not available for Sierra Leone’s nation-
al MPI, as it was first launched in 2019, but its Projections based on observed trends
global MPIT incidence fell faster than mone-
tary poverty. Before the COVID-19 pandemic, 47 coun-
tries were on track to halve multidimensional
South Asia and Sub-Saharan poverty by 2030, and 18 were off track if the
Africa lead in absolute reduction observed trends continued (figure 12).25 Of the
in multidimensional poverty 18 countries that were off track, 14 were in Sub-
Saharan Africa and were among the poorest,
As the poorest regions in the time periods suggesting that they will require a substantial
studied, South Asia and Sub-Saharan Africa boost in resources and action to halve multidi-
had the largest annualized absolute reductions mensional poverty. Results for the remaining
in multidimensional poverty (figure 11). Three 10 countries differ according to the projection
South Asian countries (Bangladesh, India and model used, though the model based on linear
Nepal) were among the 16 fastest countries to trends projects that for 9 of those countries,
reduce their MPIT value. multidimensional poverty will be halved.
12 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020FIGURE 9
In 52 of the 71 countries with both multidimensional and monetary poverty data, the incidence of multidimensional poverty fell faster in
absolute terms
Incidence of multidimensional povertya Incidence of monetary poverty ($1.90 a day)
Sierra Leone (2013–2017)
Rwanda (2010–2014/2015)
Timor-Leste (2009/2010–2016)
Sao Tome and Principe (2008/2009–2014)
Lao People's Democratic Republic (2011/2012–2017)
Mauritania (2011–2015)
Congo (2005–2014/2015)
Honduras (2005/2006–2011/2012)
Liberia (2007–2013)
Côte d’Ivoire (2011/2012–2016)
Lesotho (2009–2014)
India (2005/2006–2015/2016)
Bolivia (Plurinational State of) (2003–2008)
Bangladesh (2004–2014)
Nepal (2011–2016)
Malawi (2010–2015/2016)
Eswatini (Kingdom of) (2010–2014)
Kenya (2008/2009–2014)
Guinea (2012–2016)
Nicaragua (2001–2011/2012)
Mongolia (2010–2013)
Uganda (2011–2016)
Tanzania (United Republic of) (2010–2015/2016)
Gambia (2005/2006–2013/2014)
Zambia (2007–2013/2014)
Ghana (2011–2014)
Mozambique (2003–2011)
Zimbabwe (2010/2011–2015)
China (2010–2014)
Gabon (2000–2012)
Pakistan (2012/2013–2017/2018)
Mali (2006–2015)
Namibia (2006/2007–2013)
Sudan (2010–2014)
Burundi (2010–2016/2017)
Peru (2005–2012)
North Macedonia (2005/2006–2011)
Nigeria (2013–2016/2017)
Senegal (2005–2017)
Ethiopia (2011–2016)
Tajikistan (2012–2017)
Central African Republic (2000–2010)
Madagascar (2003/2004–2008/2009)
Cameroon (2011–2014)
Iraq (2011–2018)
Kyrgyzstan (2005/2006–2014)
Indonesia (2012–2017)
Togo (2010–2013/2014)
Burkina Faso (2006–2010)
Congo, Democratic Republic of the (2007–2013/2014)
Dominican Republic (2007–2014)
Belize (2011–2015/2016)
Egypt (2008–2014)
Niger (2006–2012)
Guyana (2009–2014)
Philippines (2013–2017)
Bosnia and Herzegovina (2006–2011/2012)
Mexico (2012–2016)
Colombia (2010–2015)
Jamaica (2010–2014)
Thailand (2012–2015/2016)
Albania (2008/2009–2017/2018)
Chad (2010–2014/2015)
Kazakhstan (2010/2011–2015)
Moldova (Republic of) (2005–2012)
State of Palestine (2010–2014)
Armenia (2010–2015/2016)
Ukraine (2007–2012)
Jordan (2012–2017/2018)
Serbia (2010–2014)
–5 –4 –3 –2 –1 0 1 2
Annualized change in headcount ratio
a. Refers to MPIT , the Multidimensional Poverty Index estimate that is based on harmonized indicator definitions for strict comparability over time.
Source: Alkire, Kovesdi, Mitchell and others 2020.
Charting pathways out of multidimensional poverty: Achieving the SDGs | 13FIGURE 10
Impact of COVID-19
Overlaying trends in the incidence of national and
international monetary and multidimensional poverty
provides a fuller picture of a country’s poverty The COVID-19 pandemic has jeopardized
situation: Colombia, Pakistan and Sierra Leone progress in reducing multidimensional poverty.
Substantial impacts on multidimensional pov-
National monetary poverty erty are anticipated through two indicators on
$3.20 a day
$1.90 a day
which the global MPI is based that are being se-
National MPI value verely affected by the pandemic—nutrition and
Global MPIT value children’s school attendance.26 This section pro-
vides simulations of multidimensional poverty
Sierra Leone if deprivation across those indicators increases
% of population to different extents.27 The analysis includes 70
90 countries with 4.8 billion people.28
The COVID-19 pandemic has interrupted
80 education globally, as schools close in the face of
national and local lockdowns. School closures
70
peaked in April 2020, with over 91 percent
60 of the world’s learners out of school. Between
May and July 2020 the proportion of learners
50 out of school fell gradually, from over 70 per-
40
cent to over 60 percent.29 In the simulations of
2000 2005 2010 2015 2020 the impact on multidimensional poverty, the
conservative scenario for school attendance
anticipates continued moderate improvements
Pakistan
over the remainder of 2020 and assumes that
% of population
50 percent of primary school–age children in
60
the countries analysed will experience contin-
ued interruption to school attendance.
40
The COVID-19 pandemic has also disrupt-
ed livelihoods and food supply chains globally.
According to the World Food Programme,
20 the number of people facing acute food inse-
curity may increase by 130 million across 55
countries.30 The simulations of the impact on
0 multidimensional poverty extend this to all 70
2005 2010 2015 2020 countries covered in the analysis, and the mod-
erate scenario for nutrition anticipates that
Colombia about 25 percent of multidimensionally poor
% of population or vulnerable people who were not undernour-
40 ished before the pandemic become undernour-
ished. In the hope that some potential rise in
30 food insecurity is prevented, or less correlated
with other deprivations, the lower-impact
20 scenario explores what would happen if about
10 percent of the already poor or vulnerable
10 but not undernourished become undernour-
ished. Conversely, recognizing that the World
0 Food Programme estimates represent only
2010 2012 2014 2016 2018 56 percent of the population of the countries
covered, in the worst-case or upper-impact
MPIT is the Multidimensional Poverty Index estimate that is based on harmonized
indicator definitions for strict comparability over time. scenario about 50 percent of the already poor
Source: Alkire, Kovesdi, Pinilla-Roncancio and Scharlin-Pettee 2020. or vulnerable but not undernourished become
undernourished.
14 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020FIGURE 11
South Asia and Sub-Saharan Africa had the largest annualized absolute reductions in multidimensional
poverty
MPIT value
0.700
Sub-Saharan Africa
South Asia
0.600 East Asia and the Pacific
Latin America and the Caribbean
Arab States
0.500 Europe and Central Asia
0.400
0.300
0.200
0.100
0.000
y1 y2 y1 y2 y1 y2 y1 y2 y1 y2 y1 y2
Year of the survey
MPIT is the Multidimensional Poverty Index estimate that is based on harmonized indicator definitions for strict comparability over time.
Source: Alkire, Kovesdi, Mitchell and others 2020.
Combining the conservative scenario of impact on livelihoods and nutrition, addition-
the impact on school attendance (50 percent) al simulations were implemented to assess the
with the moderate scenario of the impact on impact of COVID-19 on multidimensional
nutrition (25 percent), the simulations indicate poverty through just the nutrition indicator.
that the aggregate global MPI across the 70 In that case, under the moderate scenario the
countries could increase from 0.095 to 0.156 in aggregate global MPI across the 70 countries
2020, which is the same value as around 2011 could increase from 0.095 to 0.125 in 2020,
(figure 13). So, the increase in deprivations which is the same value as around 2015 (see
because of COVID-19 may set poverty levels figure 13). This increase in deprivations
back by 9.1 years, with an additional 490 mil- because of COVID-19 would set poverty re-
lion people falling into multidimensional pov- duction back by 5.2 years, with an additional
erty across the 70 countries (table 1). 237 million people falling into multidimen-
Recognizing that the impact on school sional poverty across the 70 countries (see
attendance may be less persistent than the table 1).
Charting pathways out of multidimensional poverty: Achieving the SDGs | 15FIGURE 12
Forty-seven countries are on track to halve multidimensional poverty by 2030, and eighteen are off track if observed trends continue
2030 target
Progress (logistic model)
MPIT value Linear model
Constant rate of change
0.600
Off track On track On track
(all models) (some models) (all models)
0.400
0.200
0.000
Turkmenistan
Albania
Guyana
Kyrgyzstan
Bosnia and Herzegovina
Belize
Suriname
Indonesia
Egypt
China
Dominican Republic
Chad
Niger
Burkina Faso
Ethiopia
Central African Republic
Burundi
Mali
Madagascar
Congo (Democratic Republic of the)
Benin‡
Senegal
Togo‡
Nigeria
Sudan
Cameroon‡
Pakistan
Serbia‡
Jordan‡
Mozambique
Uganda
Tanzania (United Republic of)
Gambia
Zambia
Namibia
Zimbabwe
Mexico
Colombia
Jamaica‡
Guinea
Sierra Leone
Liberia
Mauritania
Malawi
Côte d’Ivoire
Rwanda
Timor−Leste
Haiti
Kenya
Bangladesh
Cambodia
Lesotho
Nepal
India
Congo
Ghana
Sao Tome and Principe
Eswatini (Kingdom of)
Honduras
Gabon
Nicaragua
Iraq
Mongolia
Peru
Bolivia (Plurinational State of)
Tajikistan
Philippines
Thailand‡
Palestine, State of‡
North Macedonia
Moldova (Republic of)
Kazakhstan
Armenia‡
Ukraine‡
Lao People’s Democratic Republic
‡
indicates that the underlying change is not significant at p < .05.
Note: The top of the red line is the projected starting MPIT value in 2015, the dots are the projected MPIT value in 2030 and the black line is the MPIT value that would reflect multidimensional poverty being halved between
2015 and 2030. If all three dots are below the black line, a country is on track regardless of model.
Source: Alkire, Nogales and others 2020.
16 | GLOBAL MULTIDIMENSIONAL POVERTY INDEX 2020FIGURE 13
Under a conservative scenario of the impact of COVID-19 on school attendance and a moderate scenario of
the impact on nutrition, simulations indicate that the increase in deprivations because of COVID-19 may set
poverty levels back by 9.1 years, with an additional 490 million people falling into multidimensional poverty
Aggregate global Multidimensional Poverty Index value
Projected aggregate global
0.250 Multidimensional Poverty
Index value
Modelled COVID shock
Impact on nutrition (moderate
0.200 scenario) and school attendance
(conservative scenario)
Impact on nutrition
(moderate scenario)
0.150
0.100
0.050
9.1 years 5.2 years
setback setback
0.000
2005 2010 2015 2020
Note: Aggregate global Multidimensional Poverty Index projection, with simulations of setbacks in poverty reduction due to the impact of the COVID-19 pandemic.
Simulated (conservative) impact on school attendance: 50 percent of primary school–age children attending school stop attending. Simulated (moderate) impact on
nutrition: 25 percent of people who were poor or vulnerable but not undernourished become undernourished. Upper (lower) scenarios: 50 percent (10 percent) of people
who were poor or vulnerable but not undernourished become undernourished. The analysis covers 70 of the 75 countries with trends data; Colombia, Dominican Republic,
Indonesia, Philippines and Ukraine were excluded because analysis is not feasible due to missing information on the nutrition indicator.
Source: Alkire, Nogales and others 2020.
TABLE 1
COVID-19 scenarios, projected global Multidimensional Poverty Index values, increases in the number of
multidimensionally poor people, and length of setback
COVID-19 scenario Projection for 2020
Share of primary
Share of people school–age children
who are poor or who experience
vulnerablea and interruption to Increase in
become deprived in school attendance the number of
Multidimensional multidimensionally
Nutrition School attendance Poverty Index poor people Setback
(%) value (million) (years)
10 na 0.112 131 3.1
25 na 0.125 237 5.2
50 na 0.134 310 6.4
10 50 0.144 413 7.8
25 50 0.156 490 9.1
50 50 0.164 547 9.9
n a is not applicable.
a. See definition of vulnerable to multidimensional poverty in statistical table 1.
Note: Pre-COVID-19 estimates are 0.095 for MPI value and 941 million for the number of people in multidimensional poverty. The analysis covers 70 of the 75 countries
with trend data; Colombia, Dominican Republic, Indonesia, Philippines and Ukraine are excluded because of missing data for the nutrition indicator.
Source: Alkire, Nogales and others 2020.
Charting pathways out of multidimensional poverty: Achieving the SDGs | 17PART II
The Sustainable
Development Goals and the
global Multidimensional
Poverty IndexKey findings rural women face more disadvantage than
their male counterparts, and the differences
• Of the 1.3 billion multidimensionally poor by sex are higher (by about 2 years) among
people, 82.3 percent are deprived in at least the nonpoor and vulnerable groups.
five indicators simultaneously. • 84.2 percent of multidimensionally poor
• 71 percent of the 5.9 billion people covered people live in rural areas, where they are
experience at least one deprivation; however, more vulnerable to environmental shocks.
the average number of deprivations they ex- • In every developing region the proportion of
perience is five. people who are multidimensionally poor is
• There is a negative, moderate but statistically higher in rural areas than in urban areas.
significant correlation between the incidence • In Sub-Saharan Africa 71.9 percent of people
of multidimensional poverty and the coverage in rural areas (466 million people) are multi-
of three doses of the diphtheria, tetanus and dimensionally poor compared with 25.2 per-
pertussis (DTP3) vaccine. Some of the poorest cent (92 million people) in urban areas.
countries (Central African Republic, Chad, • In South Asia 37.6 percent of people in rural
Guinea, South Sudan) vaccinate less than half areas (465 million people) are multidimen-
of surviving infants with the DTP3 vaccine. sionally poor compared with 11.3 percent
• In Nigeria, which has one of the lowest per- (65 million people) in urban areas.
centages of DTP3 coverage globally, the per- • Deprivation in access to clean cooking fuel
centage of people who are poor and deprived persists worldwide: 20.4 percent of people in
in child mortality is the highest among the developing countries covered by the MPI
comparator countries Democratic Republic are multidimensionally poor and lack access
of the Congo, Ethiopia and Pakistan. This to clean cooking fuel.
suggests that child deaths can be prevented • Deprivation in access to clean cooking fuel
and multidimensional poverty reduced by among poor people in rural areas and urban
widespread immunization programmes. areas in Sub-Saharan Africa as well as in ru-
• Multidimensionally poor people have less ral areas in South Asia, the Arab States and
access to vaccinations: in the four countries Latin America and the Caribbean requires
studied, the percentage of people living with urgent attention.
a child who did not receive the third dose • Environmental deprivations are most acute
of the DPT-HepB-Hib vaccine31 is higher in Sub-Saharan Africa: at least 53.9 percent
among multidimensionally poor people and of the population (547 million people) is
people vulnerable to multidimensional pov- multidimensionally poor and faces at least
erty than among nonpoor people. one environmental deprivation. Environ-
• Sub-Saharan African countries have the mental deprivations are also high in South
highest percentages of people who are mul- Asia: at least 26.8 percent of the population
tidimensionally poor and deprived in years (486 million people) is multidimensionally
of schooling (Niger, Burkina Faso, South poor and lacks access to at least one of the
Sudan, Chad and Ethiopia) and school at- three environment indicators.
tendance (South Sudan, Burkina Faso, Niger, • There is a strong positive association between
Chad and Mali). employment in agriculture and multidimen-
• In Haiti, with the highest percentage of sional poverty, particularly in Sub-Saharan
people who are multidimensionally poor Africa. Agricultural employment may not
and deprived in years of schooling in Latin help reduce poverty in these countries with-
American and the Caribbean (22.8 percent), out additional pro-poor policy interventions.
Charting pathways out of multidimensional poverty: Achieving the SDGs | 19You can also read
