Multidimensional Measurement: Informing Policy Around the World

Sabina Alkire, September 2014 ESRC-DFID Joint Fund for Poverty Alleviation Research Grantholder Conference: Research Uptake and Impact Outline

1. Content At-a-Glance 2. Infrastructure a. Academic b. Communications c. Policy d. Relational, Opportunistic e. Ethos (language, rejection, chocolate)

MEASURING MULTIDIMENSIONAL POVERTY (Content) Motivation: Action ‘with vigour’ “positive changes have often occurred and yielded some liberation when the remedying of ailments has been sought actively and pursued with vigour”

Jean Dreze and India: An Uncertain Glory 2013 Global Multidimensional Poverty Index UNDP Human Development Report 2014 & Alkire Conconi and Seth 2014 Develop a deprivation profile for each person, using a set of indicators, cutoffs and weights. Example: .

Global Multidimensional Poverty Index UNDP Human Development Report 2014 & Alkire Conconi and Seth 2014 Identify someone as poor if he or she is deprived in 33% (for example) or more of the weighted indicators. National : you choose indicators/weights /cutoffs

Deprivation Score 33%

Aggregation: Alkire & Foster - Appropriate for Ordinal data -

The MPI is the product of two components:

MPI = H × A

1) Incidence ~ the percentage of people who are poor, or the headcount ratio H. 2) Intensity of people’s deprivation ~ the average percentage of dimensions in which poor people are deprived A.

Policy Interest – Why? 1. Intuitive – easy to understand headline 2. Birds-eye view - can be unpacked a. by region, ethnicity, rural/urban, etc b. by indicator, to show composition c. by ‘intensity’ to show inequality among poor 3. Adds Value: a. focuses on people with multiple deprivations b. shows people’s simultaneous deprivations. 4. Incentives to reach the poorest of the poor 5. Flexible you choose indicators/cutoffs/values 6. Robust to wide range of weights and cutoffs 7. Academically Rigorous – axiomatic & empirical 75% What MPI shows – National level 70% 65% How MPI decreased in Nepal 2006-11 60%

55% Nepal 2006

50% Nepal 2011 45%

40%

Average Intensity Intensity ofAverage (A) Poverty 35%

30% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Incidence - Percentage of MPI Poor People (H) Decomposition By Region (or social group) – shows inequalities How did MPI go down?

Monitor each indicator

64 Indicator Changes by region (Nepal)

Nutrition 0.03 Child Mortality 0.01 Years of Schooling Attendance -0.01 Cooking -0.03 Fuel Sanitation

-0.05 Water Annualized Absolute Change AbsoluteChange Annualized -0.07 Electricity Floor in proportion in whoandis deprived in... poor -0.09 Assets -0.11 MPI: Two kinds ~ both useful Internationally comparable: Example: The Global MPI estimated and analysed by OPHI and published by UNDP’s HDRO can be compared across 108 countries. Facilitates ‘lessons learned’ across countries. - Like $1.25/day and $2/day poverty measures & MDGs

MPI: Two kinds ~ both useful Context-Specific: Example: National MPIs reflect national contexts and priorities. They guide policies like targeting and allocation and monitor changes - Vital for policy. Not comparable: tailor-made to context.

INFRASTRUCTURE Academic Aggregation: Alkire & Foster - Appropriate for Ordinal data - The MPI is the product of two components:

MPI = H × A

1) Incidence ~ the percentage of people who are poor, or the headcount ratio H. 2) Intensity of people’s deprivation ~ the average percentage of dimensions in which poor people are deprived A.

Properties: AF Family of measures a Adjusted FGT is Ma = m(g (t)) for a > 0 Domains  0 0 0 0     0 0.42a 0 1a  g a ( k )  Persons  a a a a  0.04 0.17 0.67 1    0 0 0 0    Theorem 1 For any given weighting vector and cutoffs, the

methodology Mka =(ρk,Ma) satisfies: decomposability, replication invariance, symmetry, poverty and deprivation  focus, weak and dimensional monotonicity, nontriviality, normalisation, and weak rearrangement for a>0; monotonicity for a>0; and weak transfer for a>1. Alkire Foster JPubE 2011 Methodology Publications

• Alkire, S. and Foster, J. (2007). ‘Counting and Multidimensional Poverty Measurement’. OPHI Working Paper 7. – Presented at: Royal Economic Society, American Economic Association (AEA), International Economic Association, Econometric Society, LACEA, Social Choice and Welfare, IARIW, EADI, ECINEQ, HDCA, Development Studies Association, the World Bank, UNICEF, UNDP, IADB, Asian Development Bank, and other Universities. • Alkire, S. and Foster, J. (2011). ‘Understandings and Misunderstandings of Multidimensional Poverty Measurement’. Journal of , 9: 289– 314. – Controversial: Generated 6 academic responses, including ours to a leading critic • Alkire, S. and Foster, J. (2011). ‘Counting and Multidimensional Poverty Measurement’. Journal of Public , 95(7–8): 476–487. – Acceptable

71 Incidence and Intensity by Country . tab water hh_d_water [aw=weight], miss

| RECODE of water | (Drinking Water) Drinking Water | 0 1 | Total ------+------+------piped into dwelling | 345.27666 0 | 345.27666 piped to yard/plot | 309.02379 0 | 309.02379 public tap/standpipe | 1,342.982 0 | 1,342.982 tube well or borehole | 1,709.798 0 | 1,709.798 dug well/protected we | 280.69506 0 | 280.69506 dug well/unprotected | 0 121.95809 | 121.95809 protected spring | 1.4171675 0 | 1.4171675 unprotected spring | 0 71.644782 | 71.644782 rainwater |22.4582991 0 |22.4582991 tanker truck | 0 73.87955 | 73.87955 cart with small tank | 0 13.652633 | 13.652633 surface water (river/ | 0 474.90399 | 474.90399 bottled water | 12.62144 0 | 12.62144 other | 0 338.68866 | 338.68866 ------+------+------Total | 4,024.272 1,094.728 | 5,119

Empirical Publications

• Alkire, S. and Santos, M. E. (2013). ‘A Multidimensional Approach: Poverty Measurement & Beyond’. Social Indicators Research, 112(2): 239–257. – Empirical Play space for experimentation – junior researchers

• Alkire, S. and Santos, M. E. (2010). ‘Acute Multidimensional Poverty: A New Index for Developing Countries’. OPHI Working Paper 38, also published as Human Development Research Paper 2010/11.

• Alkire, S. and Santos, M. E. (2014). ‘Measuring Acute Poverty in the Developing World: Robustness and Scope of the Multidimensional Poverty Index’. World Development, 59: 251–274.

75 Next:

• Alkire, S., J. Foster, S. Seth, M.E. Santos, J. Roche, P. Ballon

• Multidimensional Poverty: Measurement and Analysis, OUP, 2015

• (after that we need a readable Handbook for policy)

76 INFRASTRUCTURE Communications People and Stories

Policy Briefings Infographics Interactive DataBank with Maps Data Tables Academic Paper Drafts for Comment UN General Assembly Side Event Sept 2013: Press release Over 20 governments pressure UN to change how it measures poverty

Germany, Colombia and Mexico lead calls for a new poverty measure at side-event at the UN General Assembly on the Post-2015 Development Agenda

A global network of more than 20 governments and institutions are using a side-event at the UN General Assembly on 24 September to argue for a new multidimensional poverty index to stand alongside an income poverty measure. Why? Focussing on ending income poverty alone in the post-2015 development context overlooks policies that address other aspects of being poor, such as a lack of access to healthcare, quality schooling, housing, electricity and sanitation.

Research by the Oxford Poverty and Human Development Initiative (OPHI) at Oxford University, among others, shows startling discrepancies between income poverty and multidimensional poverty, which takes into account other factors. The Multidimensional Poverty Peer Network – which was founded by Colombia, Mexico and OPHI – will use the side-event to make a case for the UN to include a multidimensional poverty index, or MPI, alongside the $1.25/day measure, to track progress towards nationally defined goals.

The MPI 2015+ would build on the global MPI published in the UN Development Programme’s flagship Human Development Reports, and would incorporate the most accurate indicators possible with new data post-2015. It would enable policymakers to identify more easily what poor people lack, and address interconnected aspects of poverty more effectively. Because it reflects improvements directly, the MPI2015+ also celebrates success and provides strong political incentives to reduce poverty. Slide title Finding ‘factoids’

Start from an idea or a controversy: • MDGs wrongly count countries not people. • Growth => higher GDP per capita but may not decrease multidimensional poverty.

Or a question: • There are more $1.25/day poor in MICS. Is it true for MPI poverty?

Finding ‘factoids’ Become very Curious about your results Play with your data. Find comparisons that are striking or unexpected Make sure factoids are 100% accurate and academically defensible.

Pakistan vs Niger

88 Pakistan vs Niger

89 Pakistan… the Bad News.

90 Design Metaphors

Fact: They are less passionate about measurement. Fact: They are less worried by details Fact: A catching image is remembered & repeated.

The MPI is like a High Resolution Lens…

The MPI is like a high resolution lens…

The MPI is like a high resolution lens…

You can zoom in

The MPI is like a high resolution lens…

You can zoom in

and see more

INFRASTRUCTURE Policy MPI in Action

Official National MPIs Colombia Mexico Bhutan Philippines

Other policy applications underway: China, Brazil, Malaysia, Indonesia, Chile, & others. Colombia’s Multidimensional Poverty Index (MPI)

Education Childhood & youth Labor Health Public utilities & conditions housing conditions Colombia 2011, 2013: MPI-Colombia 0.2 0.2 0.2 0.2 0.2 Educational Childhood & Housing & Work Health Conditions Youth Public Services

Schooling School Absence of Coverage Improved Attendance long-term Water unemploy- ment Access to Sanitation At the right Illiteracy health care level Formal given a 0.1 work necessity Access to Flooring infant Used to allocate services 0.1 0.1 Exterior resources in Walls national development Overcrowdin No Child g plan Labour

0.05 0.04 We are on are the right track Sustained poverty reduction Pobreza monetaria

60 49.7 Pobreza Multidimensioal 50 47.4 42.0 40 48.0 45.0 37.2 32,7 80 30 40.3 I P M 34.1 32.0 20 60.0 60 10 49.3 0 40 35.0

30.7

2004 2003 2005 2006 2007 2008 2009 2010 2011 2012 2014 2002 29.4 27.0 22.5 Pobreza extrema monetaria 20

0 20 17.7 16.4 1997 2003 2008 2010 2011 2012 2014 16 14.8 15.7 12.3 12 13.8 14.4 10,4 Fuente: DNP’SPSCV 10.6 8 9.5

4

0

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2014

From 2010 to 2012, 1.3 million persons overcame multidimensional poverty

To what do we owe these good results?

NDP

2010 2011 2012 2013

2010 Alerts: -Habitability -School achievement -Early childhood care -Income generation 2010- 2011 Policies: • Free tuition fees in primary and secondary education •National strategy for childhood care NDP “De cero a Siempre”

2010 2011 2012 2013

2010 Alerts: -Habitability -School achievement -Early childhood care -Income generation NDP

2010 2011 2012 2013

2011 Alerts: -Habitability -Lower rhythm of multidimensional poverty reduction 2011-2012 policies: •100,000 free housing solutions for poor NDP households

2010 2011 2012 2013

2011 Alerts: -Habitability -Lower rhythm of multidimensional poverty reduction -Low budget execution?

NDP

2010 2011 2012 2013

2012 Alerts: -Lower rhythm of income poverty reduction - Gaps of rural to urban income poverty maintained 2012-2013 policies •Improved CCT program “Más Familias en Acción “ (Higher amount and beneficiaries on rural areas) NDP Food Security Programa: Colombia Growyh Without 2010 2011 2012 Hunger 2013

2012 Alerts: -Lower rhythm of income poverty reduction - Gaps of rural to urban income poverty maintained NDP

2010 2011 2012 2013

2013 Alerts: -Social mobility -Youth unemployment 2012-2013 policies •CCT to increase human capital and youth employment NDP “Jóvenes en Acción”

2010 2011 2012 2013

2013 Alerts: -Social mobility -Youth unemployment - 4 key aspects:

1. Reflects the objectives of social policy 2. Coordinates public policy sectors 3. Monitors public policy 4. Informs Decision–making: 1. Geographic targeting 2. Programme composition 3. Graduation from CCTs

High Official Commission

Monitoring the national strategy for the reduction of poverty using official poverty measures ▪ Leaders – Presidency (Mandatory presence of the President of Colombia) – Department for Social Prosperity – National Planning Department

▪ Permanent members – Ministry of Health – Ministry of Labour – Ministry of Housing – Ministry of Agriculture – Ministry of Education – Ministry of Finance

112 Sectoral goals For accomplishing the strategy

0%-10% avance 10%-25% avance >25% avance

Línea Base Dato Dato Pobreza Análisis Goal PND 2008 2011 2012

MPI (Multidimensional Poverty) 34.7% 29.4% 27.0% 22.5% . Educational achievement (≥15 yrs) 58.8% 54.6% 53.1% 52.8% A(1) . Literacy (≥15 yrs) 14.2% 12.0% 12.1% 12.0% 5.4% 4.8% 4.1% 3.5% . School attendance (6-16) *** . No school lag (7-17) 33.4% 34.1% 33.3% 33.1% B(2) . Access to child care services (0-5) 12.1% 10.8% 9.4% 10.6% *** . Children not working (12-17) 5.5% 4.5% 3.7% 2.9% *** . Long-term unemployment 9.6% 9.1% 10.0% 9.3% C(3) *** . Formal employment 80.6% 80.4% 80.0% 74.7% . Health insurance 24.2% 19.0% 17.9% 0.5% D(4) . Access to health services 8.2% 6.6% 2.4% 8.9% *** 12.9% 12.0% 12.3% 10.9% . Access to water source . Adequate sewage system 14.1% 14.5% 12.1% 11.3% *** E(5) . Adequate floors 7.5% 6.3% 5.9% 5.6% . Adequate external walls 3.1% 3.2% 2.2% 2.1% *** . No critical overcrowding 15.7% 14.2% 13.1% 8.4% ***

*** Change 2011-2012 est. significant

FUENTE: DANE The Multidimensional Poverty Peer Network

Launched in June 2013 at with: • President Santos of Colombia • Ministers from 16 countries in person • A lecture from Professor Amartya Sen • http://www.ophi.org.uk/policy/policynetwork/

Supported by the German Federal Ministry for Economic Cooperation and Development (BMZ)

16 countries of the Multidimensional Poverty Peer Network (MPPN) at launch in 2013

Angola, Bhutan, Brazil, Chile, China, Colombia, ECLAC, Ecuador, El Salvador, Dominican Republic, , India, Iraq, Malaysia, Mexico, Morocco, Mozambique, Nigeria, OECD, the Organization of Caribbean States, OPHI, Peru, Philippines, SADC, and Vietnam

Supported by the German Federal Ministry for Economic Cooperation and Development (BMZ)

MPPN has 30 countries plus international agencies at July 2014 meeting

Supported by the German Federal Ministry for Economic Cooperation and Development (BMZ)

Colombia, Mexico, Germany, OPHI and the MPPN host a side event at the UN General Assembly 2013

Next Side Event: 25 Sept 2014, 1:15-2:30 Other Intersections: CEPAL: Constructing new LAC-MPI for the region

SADC: use MPI as a standard indicator to compare SADC members

MEDSTAT: Use MPI to compare North African countries

UNDP LAC: Forthcoming regional report on poverty will present MPIs from MPPN member countries.

Islamic Solidarity Fund for Development:

“The Fund views poverty as a multi-dimensional phenomenon, encompassing not only low income and consumption, but also low achievement in fundamental human rights including education, nutrition, primary health, water and sanitation, housing, crisis coping capacity, insecurity, and all other forms of human development.”

MPPN had a side-event at the Islamic Development Bank: June 2014

Organisation for Economic Cooperation and Development (OECD):

Move from poverty to inclusive well-being ● Create a new headline indicator to measure progress towards eradicating all forms of poverty, which could complement the current income-poverty indicator (Chapters 3 and 5).

Executive Summary, Development Cooperation Report 2013: Ending Poverty

Open Working Group Final Draft Document on SDGs: 19 July 2014

Sustainable Development Goals and targets Goal 1. End poverty in all its forms everywhere

1.1 by 2030, eradicate extreme poverty for all people everywhere, currently measured as people living on less than $1.25 a day

1.2 by 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to national definitions.

INFRASTRUCTURE Relational, Opportunistic

Open Working Group Final Draft Document on SDGs: 19 July 2014

Sustainable Development Goals and targets Goal 1. End poverty in all its forms everywhere

1.1 by 2030, eradicate extreme poverty for all people everywhere, currently measured as people living on less than $1.25 a day

1.2 by 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to national definitions.

INFRASTRUCTURE Ethos rejection~languages~chocolate OPHI at a glance • Global team: − 4-5 post-docs + director + policy director (not all FT) − 4 core staff (coordinator, comms, policy, project assistant) − Many part-time colleagues from & in many countries (India, Colombia, Mexico, Pakistan, US, S Africa, Argentina, Morocco.) • Advisors: − Sudhir Anand Purpose: To build a multidimensional − Tony Atkinson economic framework for reducing − Amartya Sen poverty grounded in people’s − Frances Stewart experiences and values.

INFRASTRUCTURE Ethos rejection~languages~chocolate

The OPHI team

MPI 2015+: Which Indicators? … (globally/nationally) & how?

First MPI indicators

Published: in Human Development Reports of UNDP