2015年-世界发展银行全球_Mapping_Subnational_Poverty_in_Zambia_109页_81mb

报告摘要

Summary of Mapping Subnational Poverty in Zambia

Core Content

This report provides a detailed analysis of subnational poverty in Zambia using the Living Conditions Monitoring Survey (LCMS) 2010 and the 2010 Census of Population and Housing. It outlines the methodology for poverty mapping and presents poverty estimates at the province, district, constituency, and ward levels. The findings are intended to support targeted poverty reduction initiatives, particularly the Social Cash Transfer Scheme (SCT) and the Sixth National Development Plan (SNDP).

Main Objectives

To map poverty at subnational levels using Small Area Estimation (SAE) techniques.
To improve the targeting of social safety net programs and public resource allocation.
To evaluate the effectiveness of poverty mapping in identifying areas with the highest poverty incidence and concentration.

Key Methodologies

Poverty Measurement Methodology

Consumption Aggregate: Used as a welfare indicator, including both food and non-food components.
Poverty Line Determination:
- Food Poverty Line: Based on the Cost of Basic Needs (CBN) approach, using a 1991 food basket updated with 2010 prices.
- Extreme Poverty Line: Set at K96,366 (2010) based on the food poverty line.
- Moderate Poverty Line: Derived by dividing the food poverty line by the food budget share (66%), resulting in K146,009.

Small Area Estimation (SAE) Methodology

The report follows a three-stage SAE approach:

Stage Zero: Ensures comparability between census and survey data by selecting matching variables and defining geographic partitions.
Stage One: Models household expenditure using survey data to estimate the relationship between socioeconomic characteristics and consumption.
Stage Two: Computes welfare measures at the census level using parameters derived from the survey and simulates error components to correct standard errors.

Key Data Sources

2010 Census of Population and Housing: Provides comprehensive demographic and geographic data.
LCMS 2010: A nationally representative household survey used to estimate consumption and poverty indicators.
Auxiliary Data: Administrative records and census data at constituency and ward levels to enhance spatial analysis.

Poverty Estimates

Province Level

National poverty incidence in 2010 was 60%, with extreme poverty at 42%.
Rural poverty was significantly higher than urban poverty (77.9% vs. 27.5%).
Poverty rates varied across provinces, with Luapula and Western, Eastern, and Northern provinces showing higher poverty levels due to their rural nature and poor infrastructure.

District Level

Poverty estimates were presented for all districts, highlighting significant differences in poverty incidence and concentration.
Districts with higher poverty rates often had limited access to basic services and infrastructure.

Constituency Level

The report ranks the 15 constituencies with the highest and lowest poverty incidence.
It also maps the concentration of the poor population across constituencies, revealing spatial disparities.

Ward Level

Poverty estimates were generated for all wards, with maps showing variations in poverty levels.
Wards within the same constituency or district exhibited significant differences in poverty rates.

Applications of Poverty Mapping

Social Cash Transfer Scheme (SCT): Poverty maps help identify priority areas for the national rollout of the SCT, targeting the poorest 500,000 households by 2016.
Resource Allocation: Supports better targeting of public resources and the design of more effective poverty reduction strategies.
Socioeconomic Correlates: Maps are used to analyze the relationship between poverty and factors such as education, employment, and access to infrastructure.

Spatial Adjustments

The report emphasizes the importance of spatial adjustments in poverty mapping, as district-level poverty rates can mask significant variation within constituencies and wards.
Spatial analysis helps in understanding local differences and improving the precision of poverty estimates.

Conclusion and Recommendations

Poverty and inequality remain high in Zambia, particularly in rural areas.
Highly disaggregated poverty data is essential for effective policy design and resource allocation.
The use of poverty maps can enhance the targeting of social programs and promote more equitable development.
Further work is recommended to refine the methodology and expand the use of poverty maps in monitoring progress and adjusting interventions over time.

Key Findings

High Poverty Levels: Zambia's national poverty incidence was 60% in 2010, with extreme poverty at 42%.
Rural-Urban Divide: Rural poverty was almost three times higher than urban poverty.
Regional Disparities: Provinces like Luapula and the Western, Eastern, and Northern provinces had higher poverty rates due to poor infrastructure and limited economic opportunities.
Socioeconomic Factors: Education, employment, and access to infrastructure are closely linked to poverty levels.
Poverty Mapping Utility: The method provides more detailed insights than national averages, aiding in targeted interventions.

References and Collaborations

The report was produced in collaboration with the Central Statistical Office of Zambia (CSO) and several international partners.
Key contributors include the World Bank team, led by Alejandro de la Fuente, and researchers from the University of Oxford and Middlesex University.
Peer reviewers and consultants provided critical feedback and guidance throughout the process.

Figures and Maps

A series of figures and maps illustrate poverty incidence and concentration across different geographic levels.
These visual tools highlight the spatial distribution of poverty and the effectiveness of the mapping approach in identifying vulnerable areas.