Effective Targeting of Anti-Poverty Programs in Indonesia
Last registered on November 20, 2013


Trial Information
General Information
Effective Targeting of Anti-Poverty Programs in Indonesia
Initial registration date
November 20, 2013
Last updated
November 20, 2013 3:53 PM EST
Primary Investigator
Harvard University
Other Primary Investigator(s)
PI Affiliation
PI Affiliation
Stanford University
PI Affiliation
World Bank
PI Affiliation
Additional Trial Information
Start date
End date
Secondary IDs
In developing countries, identifying the poor for the purposes of redistribution or social insurance programs is challenging because the government lacks reliable information about people’s incomes. In Indonesia, the Government had relied primarily upon two basic types of methods to define the poor: proxy-means testing (PMT) method and community targeting methods. The objective of this study in the near term was to help the Government of Indonesia (Statistics Indonesia – BPS in particular) in formulating better indicators to improve identifying poor households eligible for a variety of assistance programs. And in the long term, the research findings will help to inform the Government about which targeting approaches are most efficient and cost-effective, including a direct comparison of the PMT method versus community targeting methods. This research was a collaborative effort between the Indonesia’s Ministry of Planning (BAPPENAS), Central Bureau of Statistics (BPS), World Bank Office Jakarta (WBOJ) and Jameel Poverty Action Lab (J-PAL).
External Link(s)
Registration Citation
Alatas, Vivi et al. 2013. "Effective Targeting of Anti-Poverty Programs in Indonesia." AEA RCT Registry. November 20. https://www.socialscienceregistry.org/trials/97/history/536
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
We randomly assigned sub-villages to different method of targeting: Proxy Mean Testing (PMT), Community and Hybrid (combination of PMT and community). In the PMT treatment, targeting of households was based on predicting household’s consumption through data on household characteristics. In the community treatment, the residents of the neighborhood determine the list of beneficiaries through a facilitated, participatory poverty ranking exercise held at a community meeting. And the hybrid method combines the community ranking procedure with a subsequent PMT verification before households can be fully eligible.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Targeting performance and Satisfaction
Primary Outcomes (explanation)
Targeting performance: Household expenditure, community ranking; Satisfaction: fund disbursement, satisfaction, appropriateness, number of households that that should be added to the list, number of households should be subtracted from the list.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Within 3 provinces (North Sumatera, South Sulawesi and Central Java), 640 villages are selected randomly, stratifying the sample to consist of approximately 30% urban and 70% rural. For each village, we randomly selected one sub-village for the experiment. In each sub-village, an unconditional cash transfer was implemented. Each beneficiary household would receive a one time, $3 transfer. Each sub-village was randomly allocated to one of the three targeting methods (Proxy Mean Testing, Community or Hybrid). The number of households that would receive the transfer was set in advance through a geographical targeting approach, such that the fraction of households in a sub-village that would receive the subsidy was held constant, on average, across the treatment. After the beneficiaries were finalized, the funds were distributed. To publicize the lists, the program staff posted 2 copies of it in the visible locations. They also placed a suggestion box and a stack of complain cards next to the list.
Experimental Design Details
Randomization Method
Computer random number generator
Randomization Unit
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
640 villages
Sample size: planned number of observations
5,756 households
Sample size (or number of clusters) by treatment arms
209 villages got PMT method, 214 villages got community method and 217 villages got hybrid method.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Post Trial Information
Study Withdrawal
Is the intervention completed?
Intervention Completion Date
February 28, 2009, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
March 14, 2009, 12:00 AM +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
640 villages
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
5,756 households
Final Sample Size (or Number of Clusters) by Treatment Arms
209 sub villages got PMT method, 214 villages got community method and 217 villages got hybrid method
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers