Where are the poor: Region and District Poverty Estimates for Tanzania, 2012

Transcription

1 Where are the poor: Region and District Poverty Estimates for Tanzania, 2012 Blandina Kilama SK Conference Room, Umoja House Building, Ground Floor 30 June 2016

2 Outline Overview Population, GDP, and Employment Information on poverty Where are the poor? How was poverty mapping done What are the results? Concluding remarks?

3 Intro Population level, rate, structure, momentum and spatial distribution. TZ : 44,9m M: 21.8m (48.7%) F: 23m (51.3%) Source: NBS 2013 Population and Housing Census

4 Intro. Population level, rate, structure, momentum and spatial distribution. TZ : 44.9m Rural: 31.6m (70.4%) Urban: 13.3m (29.6%) TANZANIA (RURAL) TANZANIA (URBAN) TANZANIA 80+ TANZANIA Male Female Male Female Source: NBS 2013 Population and Housing Census

5 Intro... Sectoral composition of GDP selected years (current prices) Economic Activity Agriculture Industry Service 1992 NA series (as % of GDP at factor costs) 2001 NA series (as % of GDP at bp) 2007 NA series (as % of GDP at bp) ,336 1,658,275 3,406,146 2,789,853 5,690,446 7,181,357 26,380, % 48.0% 44.7% 32.9% 29.6% 26.8% 29.7% 47, ,885 1,215,091 1,638,459 4,431,057 5,406,038 18,742, % 14.2% 15.9% 19.3% 23.1% 20.2% 21.1% 116,449 1,440,356 3,161,164 4,139,962 9,076,622 12,692,496 38,388, % 41.7% 41.5% 48.8% 47.3% 47.4% 43.3% GDP 302,683 3,452,559 7,624,616 8,488,274 19,198,125 26,770,432 88,757,797 Source: National Bureau of Statistics, 1999; 2006; 2012, 2014

6 Intro... Employed Population by main activity Industry Agriculture Industry Service Total Employed Currently Employed Population (Main Activity Only) 1- Total 2- Informal 1990/91* 2000* 2006** 1990/91* 2000* 2006** 9,164,059 13,253,395 12,713,234 13,160 40,272 19, % 85.4% 76.5% 1.4% 2.8% 1.2% 445, , , , , , % 2.1% 4.3% 27.7% 17.8% 20.3% 1,279,449 1,935,538 2,560, ,543 1,143,487 1,321, % 12.5% 15.4% 70.9% 79.4% 78.5% 10,889,205 15,521,229 16,627, ,647 1,439,847 1,682,383 Source: NBS (2007), table 1 (annex); NBS (2012) National Accounts , table 3.

7 Intro - Employed Population by secondary activity Industry Agriculture/ hunting/ forestry Mining & quarry Manufacturing Construction Wholesale & retail trade Hotels & restaurants Transport/storage & communication Other community/social & personal service activities Currently Employed Population ( Secondary Activity Only) Total Informal Male Female Total Male Female Total 1,218, ,391 1,792, ,175 18, , % 12.3% 22.2% 10.7% 1.8% 6.5% 256, , , , , , % 6.4% 6.9% 18.7% 27.2% 22.7% 1,289 1,289 1,289 1, % 0.0% 0.0% 0.1% 0.0% 0.1% 625, ,099 1,121, , ,202 1,028, % 10.6% 13.9% 50.8% 45.5% 48.3% 76, , ,285 69, , , % 5.2% 4.0% 6.2% 22.6% 14.0% 51,882 3,144 55,026 31, , % 0.1% 0.7% 2.8% 0.1% 1.5% ,814 12,026 28, % 0.0% 0.0% 1.5% 1.2% 1.4% 854,801 3,013,198 3,867, ,022 16, , % 64.4% 47.9% 9.2% 1.6% 5.6% Totals 3,397,310 4,677,151 8,074,461 1,121,063 1,007,387 2,128,450 Source: Constructed by the authors using ILFS 2006, from Table C2 page 119 and Table D2 page 120

8 Current knowledge on poverty differential in Tanzania Information on consumption based poverty head count estimates are available in the Household Budget Survey reports (2011/12, 2007 and 2000/01 ) Using asset index generated from DHS as proxy for income poverty The best these two can do is to provide regional estimates given their respective sample sizes.

9 Current knowledge on poverty differential in Tanzania Population below poverty line original (HBS) Population below poverty line Adjusted estimate % std error estimate % std error Dar es Salaam Other Urban areas Rural areas Mainland Tanzania Poverty is a rural phenomenon. Source: Authors calculations using HBS 2011/12 and Census 2012

10 Where are the poor? We have seen differences between Dar es salaam, other urban and rural. Thus it is very likely that there are differences between regions and within regions. Knowledge on these lower level differences is crucial for policy development and planning.

11 Current knowledge on poverty differential in Tanzania Population below poverty line original (HBS) Population below poverty line Adjusted (HBS) estimate % std error estimate % std error Dar es Salaam Other Urban areas Rural areas Mainland Tanzania HBS is only representative for Dar es Salaam, Other urban and Rural only. To estimate poverty at lower levels there is a need to increase precision. Source: Authors calculations using HBS 2011/12 and Census 2012

12 Why Poverty Mapping Geographical variations - Visual aid Subnational data Scarce resource allocation Used for targeting and budgeting by: Policy makers NGO, FBOs and CSOs Development partners

13 How was poverty mapping done?.. We combine data from the 2011/12 HBS with the 2012 population census. This allows representative poverty estimates at levels below that of the region. Only estimates for Dar es Salaam and rural enumeration areas are provided. The estimates are results of modeling and simulations based on this modeling.

14 How was poverty mapping done? Stage 0: identify common variables in both survey and census. Stage 1: modeling of per adult equivalent consumption Stage 2: apply equation to the census data Stage 3: use GIS to map the obtained estimates

15 MODELING AND PREDICTION Stage 1: Data from household budget survey to predict exppc Regression analysis log(exppc) = β 0 + β 1 X 1 + β 2 X 2 + β n X n + U Output β 0 β 1 β 2 β n seβ Stage 2: Data from Census Calculation log(exppc) = β 0 + β 1 X 1 + β 2 X 2 + β n X n Estimated poverty and inequality for regions and districts Source: IFPRI, 2004

16 Results from Poverty Mapping

17 Dar es Salaam 5.2 Kilimanjaro Pwani Arusha Manyara 18.3 Morogoro 23.1 Katavi 23.9 Mbeya 24.3 Poverty Head count Njombe Mara Iringa Rukwa Dodoma Tanzania Simiyu Lindi Tabora Tanga Mtwara 33.9 Shinyanga 34.2 Ruvuma 34.9 Mwanza 35.3 Singida Kagera Geita 43.7 Kigoma

18 Poverty Head count Poverty Head count

19 Poverty Density Poverty Density

20 Poverty Density Poverty Density

21 Kigoma 13.8% Kigoma 5.4% Geita 12.3% Geita 4.9% Kagera 10.3% Kagera 3.9% Poverty Gap and Severity of Poverty Singida Mwanza Shinyanga Ruvuma Tabora Mtwara Tanga Simiyu Lindi Tanzania Dodoma Rukwa Mara Iringa Njombe Mbeya Katavi Morogoro Manyara Arusha 3.7% 3.1% 9.7% 9.4% 8.7% 8.5% 8.4% 8.3% 8.2% 7.4% 7.2% 6.8% 6.2% 6.2% 6.2% 6.1% 5.8% 5.6% 5.5% 5.3% Mwanza Singida Shinyanga Tabora Ruvuma Tanga Mtwara Simiyu Lindi Tanzania Mara Rukwa Dodoma Iringa Katavi Mbeya Njombe Morogoro Manyara Arusha 1.2% 1.0% 3.6% 3.6% 3.2% 3.2% 3.0% 3.0% 2.9% 2.9% 2.5% 2.5% 2.2% 2.1% 2.1% 2.0% 1.9% 1.9% 1.9% 1.8% Pwani 3.1% Pwani 1.0% Kilimanjaro 3.0% Kilimanjaro 0.9% Dar es Salaam 0.9% Dar es Salaam 0.3% 0% 2% 4% 6% 8% 10% 12% 14% 0% 1% 2% 3% 4% 5% 6%

22 Kilimanjaro Dar es Salaam Pwani Arusha Tanzania Mbeya Mtwara Iringa Inequality (GINI) Tanga Morogoro Manyara Mwanza Mara Dodoma Kagera Lindi Shinyanga Geita Njombe Tabora Ruvuma Simiyu Katavi Rukwa Singida Kigoma

23 Kilimanjaro Dar es Salaam Dar es Salaam Kilimanjaro Inequality (GINI) and Poverty Head Count Pwani Arusha Tanzania Mbeya Mtwara Iringa Tanga Morogoro Manyara Mwanza Mara Dodoma Kagera Lindi Shinyanga Geita Njombe Tabora Ruvuma Simiyu Katavi Rukwa Singida Pwani Arusha Manyara Morogoro Katavi Mbeya Njombe Mara Iringa Rukwa Dodoma Tanzania Simiyu Lindi Tabora Tanga Mtwara Shinyanga Ruvuma Mwanza Singida Kagera Kigoma 0.28 Geita Kigoma

24 Poverty Head Count 2000/01 vs. 2011/12 Dar es Salaam Kilimanjaro Pwani Arusha Manyara Morogoro Mbeya Mara Rukwa Iringa Dodoma Lindi Tabora Tanga Shinyanga Mtwara Mwanza Ruvuma Singida Kagera Kigoma HBS 2011/12 HBS 2000/ Dar es Salaam Kilimanjaro Pwani Arusha Manyara Morogoro Katavi Mbeya Njombe Mara Iringa Rukwa Dodoma Tanzania Simiyu Lindi Tabora Tanga Mtwara Shinyanga Ruvuma Mwanza Singida Kagera Geita Kigoma

25 Ranking of different indicators

26 Alternative applications of the small area estimation method Small area estimation can be used in Nutrition mapping, poverty mapping among small economic groups e.g. Children and disabled Disabled heads of household Disabled

27 Concluding remarks Poorest areas do NOT necessary have the largest number of poor people Not all indicators can be linked to poverty Include a village (EA) module in census The long form of the census should be administered to the whole population

28 Concluding remarks To better target the poor one should be aware of three things a) Who are the poor b) Where are the poor and c) Why are they poor Poverty mapping have answered the first two questions and more research is needed to answer the third one

29 Thank You REPOA 157 Mgombani Street, Regent Estate P.O. Box 33223, Dar es Salaam, Tanzania Tel: +255(0)(22) / Fax: +255(0)(22) repoa@repoa.or.tz