Poverty Mapping Efforts in Indonesia by pab13601

VIEWS: 0 PAGES: 23

									Poverty Mapping Efforts
      in Indonesia


      Asep Suryahadi


The SMERU Research Institute
      www.smeru.or.id
       Outline of Presentation
I.    Past Efforts to Map Poverty
II.   New Poverty Mapping Initiative
III. Uses of Poverty Maps
IV. Key Problems and Challenges
V.    Recommendations
  I. Past Efforts to Map Poverty
• Poverty reduction was never stated as
  development goal until 1994
• In Pelita VI, four major poverty reduction
  programs were launched
• Two major efforts to map poverty were
  initiated:
  - IDT Program
  - Family Welfare Development Program
              IDT Program
• Presidential Instruction on Disadvantaged
  Villages
• Targeting approach by classifying villages
  into poor (backward) and non-poor
• The classification was based on Podes
  (Village Potential) database
• Distribution of poor villages was strikingly
  different from distribution of poor people
     Family Welfare Program
• Managed by Family Planning Agency
  (BKKBN)
• Target households  5 welfare status: KPS,
  KS I, KS II, KS III, KS III+
• During crisis, used for targeting of social
  safety net programs
• Contributed to mistargeting of program
  beneficiaries
            Poverty Census
• To assess poverty status of all households
   more suitable for program targeting
• Too expensive  3 Provinces: Jakarta, East
  Java, South Kalimantan
• Conducted simultaneously with Population
  Census 2000
• Used indicators to determine poverty status
           Poverty Statistics
• Based on Consumption Module of
  SUSENAS, 3 yearly since 1976
• Representative at province-urban/rural
• Regional autonomy requires district level
  poverty statistics  based on Core
  SUSENAS:
  - aggregate consumption questionaire
  - no quantity and price information
II. New Poverty Mapping Initiative
• A new method combining detailed information
  from household survey and complete coverage
  of population census  simulated-welfare
  mapping
• Two stages:
  - using survey data, estimate correlation pattern
  - using census data, used the estimated pattern
      to predict consumption
Simulated Welfare Mapping Method
             Econometrics


  Census

                            For census household:
                              predict per capita
                               expenditure and
                                 error margin

   Survey
            The Pilot Study
• The new poverty mapping method was
  introduced in a seminar at BPS in June 2001
• BPS, SMERU, and World Bank collaborate
  in an effort to apply the method
• Two phases:
  - Pilot study of 3 Provinces: East
      Kalimantan, Jakarta, East Java
  - Application to the rest of provinces
             Data Sources
•   Consumption Module SUSENAS 1999
•   Core SUSENAS 1999
•   Population Census 2000
•   Podes 1999
      Implementation Procedure
1.   Matching variables in survey and census
2.   Selecting explanatory variables
3.   Estimating the model
4.   Simulations on census data
5.   Calculation of poverty indicators
Results: Successful Replication
            Estimates of Headcount Poverty Rates
                              Poverty Rate Standard Error
     Province & Method
                                   (%)           (%)
Jakarta:
SUSENAS                            2.82          0.62
Simulated-Welfare Mapping          2.98          0.53
East Java:
SUSENAS                           33.34          1.24
Simulated-Welfare Mapping         32.10          1.31
East Kalimantan:
SUSENAS                           21.05          3.38
Simulated-Welfare Mapping         20.52          2.35
Poverty Maps of
East Kalimantan
The Importance of Error Margin
           Urban East Kalimantan

               Provincial headcount




    40



                                      (headcounts with 2se error bounds)
    30




    20




    10




     0


         0.0       0.2                0.4        0.6        0.8
                  Predicted headcount estimates
     Precision of the Estimates

                  Mean of Standard Error
   Region Level   Jakarta       East Java   East Kalimantan
Province          0.1765         0.0408          0.1147
District          0.2678         0.1165          0.1873
Subdistrict       0.6298         0.2267          0.2552
Village           1.2796         0.5501          0.5282
Standard Error & Population Size
                                                 East Kalimantan Urban Villages

                  0.14




                  0.12




                  0.10
 Standard Error




                  0.08




                  0.06




                  0.04




                  0.02




                  0.00
                         0   2,500   5,000   7,500    10,000     12,500     15,000   17,500   20,000   22,500   25,000
                                                          Number of Population
       III. Uses of Poverty Map
•   Capturing heterogeneity of poverty
•   Improving targeting of interventions
•   Articulating policy objectives
•   Communicating distribution of welfare
•   Evaluating impact of programs
•   Incorporation into GIS
Benefit Relative to Other Methods
•   Higher resolution poverty maps
•   Based on direct measures of welfare
•   Provide measure of precision
•   Use existing data
    Promotion of Poverty Map
• Easy access
• Seminar and workshop
• Application to other welfare indicators
    Poverty Map Sustainability
• Initial production is externally driven
• Internalizing needs for poverty maps
 IV. Key Problems & Challenges
• Limited technical expertise
• Integration of BPS & other institutions data
• Perception of usefulness of poverty maps
         V. Recommendations
•   Training and workshop
•   Networking
•   Facilitating data integration
•   Supporting upstream and downstream
    research

								
To top