Oxford and Human Development Initiative (OPHI) www.ophi.org.uk Oxford Dept of International Development, Queen Elizabeth House, University of Oxford Country Briefing: Tanzania Multidimensional Index (MPI) At a Glance December 2011 This Country Briefing presents the results of the Multidimensional Index (MPI) and explains key findings graphically. Further information as well as international comparisons are available at www.ophi.org.uk/policy/multidimensional-poverty-index/. The MPI was constructed by OPHI for UNDP s 2011 Human Development Report (http://hdr.undp.org/en/). Citation: Alkire, Sabina; Jose Manuel Roche; Maria Emma Santos & Suman Seth (2011). Tanzania Country Briefing. Oxford & Human Development Initiative (OPHI) Multidimensional Index Country Briefing Series. Available at: www.ophi.org.uk/policy/multidimensional-poverty-index/mpi-country-briefings/. For more information on the MPI please see Alkire, Sabina and Maria Emma Santos. Acute Multidimensional : A New Index for Developing Countries OPHI Working Paper 38 and the latest MPI resources online: http://www.ophi.org.uk/policy/multidimensional-poverty-index/mpiresources/. Inside the MPI The MPI has three dimensions and 10 indicators, which are shown in the box below. Each dimension is equally weighted, each indicator within a dimension is also equally weighted, and these weights are shown in brackets within the diagram. Country Profile Tanzania-DHS-2008 Country: 3 Tanzania 93 Year: 2008 Survey: DHS Region: Sub-Saharan Africa Multidimensional Index (MPI) 1 1 The MPI reflects both the incidence or headcount ratio (H) of poverty the proportion of the population that is multidimensionally poor and the average intensity (A) of their poverty the average proportion of indicators in which poor people are deprived. The MPI is calculated by multiplying the incidence of poverty by the average intensity across the poor (H*A). A person is identified as poor if he or she is deprived in at least one third of the weighted indicators. The following table shows the multidimensional poverty rate (MPI) and its two components: incidence of poverty (H) and average intensity of deprivation faced by the poor (A). The first and second columns of the table report the survey and year used to generate the MPI results. Those identified as MPI poor are deprived in at least 33% of weighted indicators. Those identified as "Vulnerable to " are deprived in 20% - 33% of weighted indicators and those identified as in "Severe " are deprived in over 50%. Survey Year Multidimensional Index (MPI = H A) Incidence of (H) Average Intensity Across the Poor (A) Population Vulnerable to Population in Severe DHS 2008 0.367 65.2% 56.3% 23.0% 43.7% www.ophi.org.uk Page 1
Comparing the MPI with Other Measures Column chart A compares the poverty rate using the MPI with three other commonly used poverty measures. The height of the first column denotes the percentage of people who are MPI poor (also called the incidence or headcount ratio). The second and third columns denote the percentages of people who are poor according to the $1.25 a day income poverty line and $2.00 a day line, respectively. The final column denotes the percentage of people who are poor according to the national income poverty line. The table on the right-hand side reports various descriptive statistics for the country. The statistics shaded in khaki/olive are taken from the year closest to the year of the survey used to calculate the MPI. The year is provided below each column in chart A. Proportion of Poor People 100.0% 90.0% 80.0% 70.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% 65.2% A. Comparative Measures 67.9% 87.9% 0.367 65.2% MPI (H) US$1.25 a US$2 day a daynational Line Average Intensity of Deprivation (A) 56.3% 65% 68% 88% 33% 67.9% Income Poor ($2.00 a day) 87.9% Poor (National Line) 33.4% MPI (H) US$1.25 a day US$2 a day National Line 2008 2007 2007 2007 HDI category* Low Measure Comparing the MPI with Other Measures 33.4% Summary Multidimensional Index MPI Poor (H) Income Poor ($1.25 a day) Human Development Index 2011* HDI rank* The World Bank (2011). World Development Indicators. Washington, DC. * UNDP (2011). "Human Development Report", Statistical Table 1. New York. Note: For population figures and numbers of MPI poor people, consult the tables on OPHI s website: http://www.ophi.org.uk/policy/multidimensional-poverty-index/ Column chart B shows the percentage of people who are MPI poor (also called the incidence or headcount) in the 109 developing countries analysed. The column denoting this country is dark, with other countries shown in light grey. The dark dots denote the percentage of people who are income poor according to the $1.25 a day poverty line in each country. The graph above tells you the year this data comes from. Dots are only shown where the income data available is within three years of the MPI survey year. 0.466 152 Poor People 100% B. Headcounts of MPI Poor and $1.25/day Poor 90% 80% 70% 60% 50% 40% 30% 20% 0% Niger Ethiopia Mali Central African Republic Burundi Liberia Burkina Faso Guinea Somalia Rwanda Mozambique Angola Sierra Leone Comoros DR Congo Uganda Malawi Benin Timor Leste Senegal Madagascar Tanzania Nepal Zambia Chad Mauritania Cote d'ivoire Gambia Bangladesh Haiti Togo Nigeria India Cameroon Yemen Cambodia Pakistan Kenya Lao Swaziland Republic of Congo Zimbabwe Namibia Gabon Lesotho Sao Tome and Principe Honduras Myanmar Ghana Vanuatu Djibouti Nicaragua Bhutan Guatemala Indonesia Bolivia Peru Viet Nam Tajikistan Mongolia Iraq Philippines Guyana South Africa Paraguay China Morocco Suriname Estonia Turkey Egypt Trinidad and Tobago Belize Syrian Arab Republic Colombia Sri Lanka Azerbaijan Maldives Kyrgyzstan Dominican Republic Hungary Croatia Mexico Czech Republic Argentina Tunisia Brazil Jordan Uzbekistan Ecuador Ukraine Macedonia Moldova Uruguay Thailand Latvia Montenegro Occupied Palestinian Territories Albania Russian Federation Armenia Serbia Bosnia and Herzegovina Georgia Kazakhstan United Arab Emirates Belarus Slovakia Slovenia MPI Poor 22 22 Income Poor (living on less than $1.25 a day) www.ophi.org.uk Page 2
Incidence of Deprivation in Each of the MPI Indicators The MPI uses 10 indicators to measure poverty in three dimensions: education, health and living standards. The bar chart to the left reports the proportion of the population that is poor and deprived in each indicator. We do not include the deprivation of non-poor people. The spider diagram to the right compares the proportions of the population that are poor and deprived across different indicators. At the same time it compares the performance of rural areas and urban areas with that of the national aggregate. Patterns of deprivation may differ in rural and urban areas. C. Deprivations in each Indicator D. the Population MPI Poor and Deprived Living Standards Health Education School Attendance Drinking. Water Drinking Water 80.0% 70.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% School Attendance 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 70.0% the Population who are MPI poor and deprived in each indicator National Urban Rural Composition of the MPI The MPI can be broken down to see directly how much each indicator contributes to multidimensional poverty. The following figure shows the composition of the MPI using a pie chart. Each piece of the pie represents the percentage contribution of each indicator to the overall MPI of the country. The larger the slice of the pie chart, the bigger the weighted contribution of the indicator to overall poverty. E. Contribution of Indicators to the MPI 8% Drinking Water 7% 6% 6% School Attendance 11% 32% School Attendance Drinking Water Education Health Living standards 0% www.ophi.org.uk Page 3
Decomposition of MPI by Region The MPI can be decomposed by different population subgroups, then broken down by dimension, to show how the composition of poverty differs between different regions or groups. On the left-hand side of column chart F, the height of each of the three bars shows the level of MPI at the national level, for urban areas, and for rural areas, respectively. Inside each bar, different colours represent the contribution of different weighted indicators to the overall MPI. On the right-hand side of column chart F, the colours inside each bar denote the percentage contribution of each indicator to the overall MPI, and all bars add up to 100%. This enables an immediate visual comparison of the composition of poverty across regions. F. Contribution of Indicators to the MPI at the National Level, for Urban Areas, and for Rural Areas 0.450 100% YS, 6.1% YS, 4.4% YS, 6.3% 0.400 YS 90% SA, 10.9% SA, 10.1% SA, 11.0% 0.350 YS SA 80% MPI Value 0.300 0.250 0.200 0.150 0.100 0.050 0.000 SA CM CM No data on N No data on N YS E SA E S S CM DW DW F No data on N E S F CF A DW F CF A CF A National Urban Rural Percentage Contribution to MPI 70% 60% 50% 40% 30% 20% 0% CM, 32.3% CM, 30.9% CM, 43.2% No N, 0.0% E, 9.5% S, 9.7% DW, 7.2% F, 8.4% No N, 0.0% E, 8.0% S, 10.0% DW, 4.2% F, 4.6% No N, 0.0% E, 9.7% S, 9.6% DW, 7.5% F, 8.9% CF, 9.8% CF, 10.7% CF, 9.7% A, 6.1% A, 4.8% A, 6.3% National Urban Rural YS = CM = E = DW = Drinking Water CF = SA = School Attendance N = Nutrition S = F = A = Intensity of Multidimensional Recall that i) a person is considered poor if they are deprived in at least one third of the weighted indicators and ii) the intensity of poverty denotes the proportion of indicators in which they are deprived. A person who is deprived in 100% of the indicators has a greater intensity of poverty than someone deprived in 40%. The following figures show the percentage of MPI poor people who experience different intensities of poverty. The pie chart below breaks the poor population into seven groups based on the intensity of their poverty. For example, the first slice shows deprivation intensities of greater than 33% but strictly less than 40%. It shows the proportion of poor people whose intensity (the percentage of indicators in which they are deprived) falls into each group. The column chart H reports the proportion of the population in a country that is poor in that percentage of indicators or more. For example, the number over the 40% bar represents the percentage of people who are deprived in 40% or more indicators. 70%-79.9% 33% 40% 50% 60% 70% 80% 90% 100% H. People Deprived in X% 80%-89.9% per 0.652 90%-100% 0.517 0.437 0.304 0.128 0.060 0.016 0.010 or more of the MPI Weighted Indicators 70.0% 65.2% 0.348 0.483 0.563 0.696 0.872 0.940 0.984 0.990 40%-49.9% 60%-69.9% MPI Poor 33%-39.9% 40%-49.9%50%-59.9%60%-69.9%70%-79.9%80%-89.9% 51.7% 90%-100% 33%-39.9% 50.0% 0.136 0.080 0.133 0.176 0.068 0.044 43.7% 0.006 40.0% 30.4% 30.0% 50%-59.9% G. Intensity of Deprivation Among MPI Poor 20.0% 10.0% 0.0% 12.8% 6.0% 1.6% 1.0% 33% 40% 50% 60% 70% 80% 90% 100% Intensity of www.ophi.org.uk Page 4
Multidimensional at the Sub-national Level In addition to providing data on multidimensional poverty at the national level, the MPI can also be 'decomposed' by sub-national regions to show disparities in poverty within countries. This analysis can be easily performed when the survey used for the MPI is representative at the sub-national level. The following table shows the MPI value and its two components at the sub-national level: the incidence of poverty (H) and the average intensity of deprivation faced by the poor (A). The last two columns present the percentage of the population vulnerable to multidimensional poverty and living in severe poverty, respectively. Regional population figures, in the second column, are estimated using the weighted sample share of each region and the 2008 population estimates from UNDESA, Population Division (2011), World Population. The map shows visually how the MPI varies across regions - a darker colour indicates higher MPI and therefore greater poverty. I. Multidimensional across Sub-national Regions Region Percentage of Population Multidimension al Index (MPI = H A) Incidence of (H) Average Intensity Across the Poor (A) Population Vulnerable to Population in Severe Arusha 3.9% 0.331 58.1% 56.9% 25.1% 41.2% Dar es Salaam 6.2% 0.178 35.1% 50.7% 17.8% 19.6% Dodoma 4.1% 0.402 69.1% 58.2% 23.2% 51.5% Iringa 4.6% 0.327 62.5% 52.4% 29.5% 36.9% Kagera 5.8% 0.452 76.5% 59.2% 18.1% 58.4% Kigoma 4.3% 0.356 70.0% 50.8% 27.0% 37.8% Kilimanjaro 4.1% 0.139 29.8% 46.8% 36.4% 13.0% Lindi 2.5% 0.409 73.1% 55.9% 20.9% 47.2% Manyara 3.0% 0.342 65.7% 52.1% 28.7% 40.6% Mara 4.0% 0.381 63.5% 24.5% 50.6% Mbeya 6.3% 0.355 60.8% 58.4% 26.0% 42.3% Morogoro 4.7% 0.333 60.5% 55.1% 25.7% 40.2% Mtwara 3.3% 0.442 81.4% 54.3% 16.7% 53.5% Mwanza 9.0% 0.398 68.3% 58.3% 23.3% 51.0% Pemba 1 0.6% 0.413 70.2% 58.9% 18.6% 46.3% Pemba 2 0.5% 0.327 56.6% 57.7% 23.9% 37.6% Pwani 2.2% 0.364 67.3% 54.2% 23.4% 42.7% Rukwa 3.7% 0.473 76.1% 62.2% 19.2% 58.9% Ruvuma 3.8% 0.338 60.6% 55.8% 32.7% 42.7% Shinyanga 8.9% 0.458 79.6% 57.6% 17.9% 52.9% Singida 2.5% 0.379 73.9% 51.4% 21.3% 37.5% Tabora 5.9% 0.487 83.5% 58.3% 12.7% 52.8% Tanga 4.4% 0.363 64.8% 56.0% 25.2% 43.3% Unguja 1 0.4% 0.303 54.6% 55.4% 29.5% 32.1% Unguja 2 0.3% 0.222 41.1% 54.0% 29.4% 24.8% Unguja 3 1.2% 0.169 33.5% 50.4% 16.9% 16.1% www.ophi.org.uk Page 5
J. Mapping Rates at the Sub-national Level H_hh_assets_depr H_electricity_depr H_toilet_mdg_depr H_water_mdg2_depr H_floor_2_depr H_cooking_mdg_depr H_hh_no_dead_children_depr H_hh_nutrition_depr H_hh_all_child_enrol_depr H_hh_years_edu5_depr 50% 89% 95% 59% 73% 99% 36% 0% 25% 13% Ur_H2 Ur_M02Ur_A2 Ru_H2 Ru_M02 Ru_A2 contr_ur_h2 contr_ru_h2 contr_ur_m02 contr_ru_m02 0.39 0.20 0.51 0.72 0.41 0.57 13% 87% 12% 88% The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by OPHI or the University of Oxford. This map is intended for illustrative purposes only. Attendance Drinking WaterCooking Fuel 13% 24% 36% 63% 64% 47% 56% 65% 41% yeedu5_cenurh2 chilenrol_cenurh2 mort_cenurh2 nutr_cenurh2 electr_cenurh2 toilet_cenurh2 water_cenurh2 floor_cenurh2 cookfuel_cenurh2 assets_cenurh2 Education Health 5% 12% 26% 29% 36% 15% 17% 38% 17% 9% 26% 25% yeedu5_cenruh2 chilenrol_cenruh2 mort_cenruh2 nutr_cenruh2 electr_cenruh2 toilet_cenruh2 water_cenruh2 floor_cenruh2 cookfuel_cenruh2 assets_cenruh2 Education Health 16% 27% 38% 72% 72% 56% 66% 72% 47% 22% 38% 64% Living Standard Urb Living Standard rural