Multidimensional poverty measurement improves policy-making
Submitting InstitutionUniversity of Oxford
Unit of AssessmentAnthropology and Development Studies
Summary Impact TypeSocietal
Research Subject Area(s)
Medical and Health Sciences: Public Health and Health Services
Economics: Applied Economics, Econometrics
Summary of the impact
Poor people define poverty to include a simultaneous lack of education,
health, housing, mployment and income, among other factors. Recognising
this, Sabina Alkire and James Foster developed an axiomatic methodology of
measurement that incorporates multiple dimensions of poverty — the Alkire
Foster method (AF). The AF method provides a robust, `open-source'
measurement tool for policy-making. One key impact is an AF index covering
100+ countries, published annually in the UNDP's Human Development
Reports. Another is national adoption by three governments and a
multidimensional poverty peer network of 22 governments and agencies. The
AF method is also incorporated into other internationally recognised
well-being measures such as USAID's 19-country Women's Empowerment in
Agriculture Index, and Bhutan's Gross National Happiness index.
In 2007, Sabina Alkire established the Oxford Poverty and Human
Development Initiative (OPHI) (www.ophi.org.uk)
at the University of Oxford and, with James Foster, developed a new,
multidimensional poverty measurement methodology that is implemented by
choosing context- specific variables that go beyond income and consumption
[see Section 3: R1-R3].
The Alkire Foster (AF) method employs a set of user-designed indicators,
cut-offs and weights to identify who is poor and to generate a rigorous
class of poverty measures. The simplest measure reflects both the
incidence of poverty (the percentage of the population who are poor), and
the intensity of poverty (the share of deprivations suffered by each
household); other measures also reflect the depth and severity of poverty.
The AF method identifies who is poor by considering the joint
deprivations they experience. It then aggregates that information to
reflect societal poverty in a way that is robust and decomposable; for
example by gender, ethnicity or geographic region. The final measures can
be broken down by indicator, after identification, to show the
composition. It is a very flexible methodology; different dimensions (e.g.
nutrition) and indicators (e.g. stunting) can be used, depending on the
requirements of the policy-makers concerned, to target support to the very
Following the development of the AF method, a number of papers were
published by OPHI in which authors used the new methodology to analyse
multidimensional poverty in countries and continents around the world; for
example, in Bhutan, India and Latin America [R4, R5]. These papers
demonstrated how policy insights arise from the new methodology, by
showing how deprivations vary between groups of people and regions.
Methodological papers developed applications such as targeting, and tools
such as standard errors. A manual, Multidimensional Poverty
Measurement and Analysis, will be published by Oxford University
Press in 2014.
By comparing the new multidimensional measures with standard income
measures of poverty, this research drew attention to how the AF measures
complement income poverty statistics. In particular, the AF measure
provides a meaningful at-a-glance number, which can then be broken down to
reveal how many are poor, how they are poor, and who are the poorest [R6].
The fact that the AF measure is both compact and transparent makes it
attractive to policy-makers wishing to coordinate policy, target resources
more effectively and monitor the impact of programmes over time [R7].
- Sabina Alkire, Director, OPHI (2007-present)
- James Foster, OPHI Research Associate (2007-present), OPHI Visiting
Fellow (2006-07), and Professor of Economics and International Affairs,
George Washington University
- José Manuel Roche, OPHI Research Officer (2009-present)
- Maria Emma Santos, OPHI Research Associate (2010-present), OPHI
Research Officer (2008-10) and Universidad Nacional del Sur and CONICET,
References to the research
[R1] Alkire, S and J E Foster (2011) `Counting and
Multidimensional Poverty Measurement,' Journal of Public Economics
95 (7-8): 476-87. (Impact factor: 1.520; citations in Google Scholar:
This paper was originally published as OPHI Working Paper No. 7 (2007).
[R2] Alkire, S, and J E Foster (2011) `Understandings and
Misunderstandings of Multidimensional Poverty Measurement,' Journal of
Economic Inequality 9 (2): 289-314.
One of the top five most downloaded documents from the Journal of
[R3] (2013) Journal of Economic Inequality 9 (3): 479-99
This issue of the journal featured six thought pieces on the AF method by
Jacques Silber, David Roodman, Erik Thorbecke, Nancy Birdsall, Francisco
Ferreira, Jeni Klugman and Francisco Rodríguez and Hyung-Jin Choi. It also
included Alkire, S, J E Foster and M E Santos `Where Did Identification
Go?' (pp. 501-5).
[R4] Alkire, S and M E Santos (eds) (2013) `A Multidimensional
Approach: Poverty Measurement and Beyond'. Special issue of Social
Indicators Research 112 (2). (Impact factor: 1.264.)
[R5] Alkire, S., and M.E. Santos (2010) `Acute Multidimensional
Poverty: A New Index for Developing Countries,' UNDP Human Development
Report Research Paper, 2010/11
Presented on request to the DFID Chief Economist's Seminar and the Poverty
and Inequality Measurement and Analysis Practice Group of the World Bank.
[R6] Alkire, S, R Meinzen-Dick, A Peterman, A R Quisumbing, G
Seymour and A Vaz (2013) `The Women's Empowerment in Agriculture Index', World
Development 52: 71-91. (Impact factor: 1.527.)
Key research awards that supported this and related research:
• IDRC-CIDA Project 104071, `Human Development and Capabilities Network -
Phase III', $2,107,500 (CDN), 2006-10
• ESRC-DFID Grant Reference ES/I032827/1, `Multidimensional Poverty:
Enriching Methodologies of Measurement and Policy Analysis', £387,416,
• UNDP Human Development Report Office grants for developing the
Multidimensional Poverty Index and training programme for policy-makers,
$326,572 (USD), 2009-12
• UK Department for International Development (DFID) Grant Component Code
200440-102, `Multidimensional Poverty: Consolidating Innovation', £94,305,
• USAID-IFPRI grant to create the Women's Empowerment in Agriculture
Index, $124,450 (USD), 2011-12
• BMZ-Georg-August-Universität Göttingen grant on `Multidimensional
Poverty Measurements: On Developing Alternative Poverty Measures for
Development Policy', €140,300, 2011-13
• BMZ-GIZ Project 11.2066.6-002.00, `Millennium Goals and Poverty
Reduction', seed grant for a multidimensional poverty peer network for
policy-makers, €100,000, 2012-13
Details of the impact
Since its first publication in working paper form in 2007, the AF method
has been adapted by several governments to create tailored
multidimensional measures of poverty and well-being which contribute
directly to poverty reduction [see Section 5: C1-C5]. Furthermore,
it is used to construct the 104-country Multidimensional Poverty Index
(MPI) included in UNDP's annual flagship publication Human Development
Report (HDR) [C6], and in the monitoring and evaluation of
US government aid programmes [C7, C8].
In June 2013, OPHI with President Santos of Colombia and high-level
representatives of 22 countries, the Organisation for Economic Cooperation
and Development (OECD), the Economic Commission for Latin America and the
Caribbean (ECLAC) and the Southern African Development Community (SADC),
launched the Global Multidimensional Poverty Peer Network. The Network
connects policy-makers who are exploring or implementing official
multidimensional poverty measures based on the AF method at the regional,
national and international levels. Network participants include China,
India, Nigeria, Brazil, Philippines, Iraq, and Morocco, and OPHI acts as
the Network's Secretariat.
UNDP Human Development Report Office (HDRO) — Global Multidimensional
OPHI collaborated with the HDRO to develop a new
Multidimensional Poverty Index in 2009-10. This index was constructed by
Alkire and Santos using the AF method, was launched in the Human
Development Report in 2010 and updated with new estimations and
analyses for subsequent reports [C6]. Covering over 100 countries,
the MPI shows the nature and extent of poverty from the household up to
the regional, national and international level. It continues to attract
attention from the media and high-profile thinkers and commentators around
the world, including, in 2013, the Economist, the Telegraph and the
Observer in the UK, the China Post, Bloomberg, Fox News, and the Times of
Government of Mexico — National Multidimensional Poverty Measure
Following a 2006 law, requiring the construction of a multidimensional
poverty measure, the Mexican government's National Council for the
Evaluation of Social Policy examined five multidimensional poverty
measurement methodologies, including the AF method. Based on their
research, Alkire and Foster participated as international experts in a
two-year process, applying the AF model to suit the specifications of the
Mexican government via a two-way dialogue process. Mexico's official
multidimensional poverty measure, using a version of the AF method, was
launched in 2009 and updated in 2011 and 2013. This pioneering measure
reflects people's simultaneous deprivations in income and social rights,
including health, housing, education and food, enabling policy-makers to
devise more effective poverty-reduction strategies [C1].
Government of Colombia — National MPI and Poverty-Reduction Strategy
In 2011, the government of Colombia launched a national multidimensional
poverty measure using the AF method and developed in close collaboration
with OPHI researchers. Devised by Colombia's Ministry of Planning, the
measure reflects the national plan's goals and targets. It has been
cascaded through the policy process, and is used for policy coordination,
for geographical and household targeting, and for monitoring and
evaluation. The MPI is now an official statistic, updated annually based
on new survey data [C2].
Government of Bhutan — Gross National Happiness (GNH) Index
In 2008, the president of the government of Bhutan's Centre for Bhutan
Studies (CBS) co-developed the first (pilot) GNH Index with Alkire using
the AF method, and co-authored two OPHI working papers. The index was
updated and strengthened in 2012, again using the AF method, with Alkire
as co-author [C3]. The landmark measure has nine domains and 33
indicators, and the surrounding policy framework creates incentives for
the government, as well as NGOs and businesses, to increase GNH. Bhutan
has also adapted the AF method to construct a national MPI, with tailored
indicators that reflect deprivations in rudimentary services and core
USAID and IFPRI — Women's Empowerment in Agriculture Index (WEAI)
Alkire and OPHI consultant Ana Vaz constructed the WEAI in 2012 in
partnership with the US government's Feed the Future initiative, the
United States Agency for International Development (USAID) and the
International Food Policy Research Institute (IFPRI). The WEAI, which was
commissioned by the US government to monitor the impact of its Feed the
Future interventions, captures women's empowerment in the agricultural
sector directly, without using proxies such as income and education [R6].
Alkire and Vaz developed the index from pilot survey data using an
adaptation of the AF method. The work has now been expanded to
representative datasets in 19 countries, and there is uptake from NGOs,
including CARE and Oxfam, and interest from the UN's Food and Agriculture
Organisation (FAO) in further implementations [C7, C8].
In addition to the users above, the AF method has been adapted and
applied at the national or subnational level in China [C4],
Malaysia, and Minas Gerais and São Paulo in Brazil [C5]. Official
National AF multidimensional poverty measures are under development in
Chile [C9], El Salvador, and Vietnam among others, and tailor-made
AF MPIs have been published in UNDP national and regional Human
Development Reports in Mercosur, Eastern Europe and Central Asia,
Malaysia, Nicaragua and Iraq, among others [C10-C12], often
sparking consideration as national indices.
Further impact occurred through OPHI's training of policy-makers, which
helped to build their technical capacity to implement an AF measure.
Two-week training courses have been held in Chile, Nicaragua, India,
Indonesia, Jordan, the Netherlands, Peru, the UK and the US, and tailor-
made courses given in locations such as Bhutan, Brazil, Egypt, Ethiopia,
Hungary, Jordan, Lebanon, Malaysia, Rwanda, Saudi Arabia, Thailand, the US
Sources to corroborate the impact
[C1] CONEVAL (2010) A
Methodology for the Measurement of Multidimensional Poverty.
Mexico City: CONEVAL (see also http://www.ophi.org.uk/policy/national-policy/mexico-mpi).
[C2] Angulo Salazar, R.C., Y. Díaz Cuervo, and R. Pardo Pinzón
de Pobreza Multidimensional para Colombia (IPM-Colombia) 1997-2010.
Bogota: Republic of Colombia Department of Planning (see http://www.dnp.gov.co/
for other sources)
[C3] Alkire, S, K Ura, T Zangmo, and K Wangdi (2012) An
Extensive Analysis of GNH Index. Thimphu: Centre for Bhutan
Studies. UNDP (2011) Bhutan National Human Development Report 2011:
Progress: Rising to the Climate Challenge mentions Bhutan's
national MPI (pp 31-33).
[C4] Xiaolin Wang (2013) Developing
GIS of the National Poverty Reduction in China. Presentation
given at the launch of the Multidimensional Poverty Peer Network, Oxford,
June 2013, describing the IPRCC's work to identify and monitor
multidimensional poverty using the AF method.
Travessia - Sobre o programa. This webpage explains that the
Travessia poverty- reduction programme introduced by the Minas Gerais
state government in Brazil uses the Multidimensional Poverty Index
developed by OPHI and the UNDP's HDRO.
[C6] The UNDP Human Development Report Office: corroborates the
use of the AF method to develop the global MPI published in the 2010 and
subsequent Human Development Reports (held on file).
[C7] The Monitoring and Evaluation Team, Food Security, USAID:
corroborates the use of the AF method to construct the Women's Empowerment
in Agriculture Index (held on file).
[C8] USAID, Feed the Future, IFPRI and OPHI (2012) The
Women's Empowerment in Agriculture Index. Washington DC: IFPRI
[C9] Comisión para la Medición de la Pobreza (2013) Informe
de Avance, abril 2013. This report states that Chile
should incorporate multidimensional poverty measurement and the dimensions
that should be used (see Section V).
[C10] UNDP (2011) Las
juventudes construyendo Nicaragua. This
Spanish-language report, which won a 2013 Human Development Award for
Excellence, describes the Multidimensional Poverty Youth Index (IPMJ),
which was constructed using the AF method (see p 77 onwards).
[C11] UNDP (2009) Human Development Report for Mercosur,
para incluir: jóvenes y desarrollo humano. This report
describes a MPI for young people in Mercosur, the trading bloc set up by
Argentina, Brazil, Paraguay and Uruguay.
[C12] UNDP (2011) Regional Human Development Report,
Europe and Central Asia: Beyond
Towards Inclusive Societies. This report presents a
Multidimensional Social Exclusion Index based on the AF method.