Multidimensional Poverty Measurement: Informing Policy Around the World
Total Page:16
File Type:pdf, Size:1020Kb
Multidimensional Poverty 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 Amartya Sen 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 Absolute Change Annualized -0.07 Electricity Floor in proportion in whoand is 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 Economic Inequality, 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 Economics, 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