Does Elite Capture Matter? Local Elites and Targeted Welfare Programs in Indonesia (reproducible scripts)
2019 - IDN_2019_ECTWP_v01_RR
Vivi Alatas, Abhijit Banerjee, Rema Hanna, Benjamin A. Olken, Ririn Purnamasari, Matthew Wai-Poi
Overview
Abstract
This paper investigates how elite capture affects the welfare gains from targeted government transfer programs in Indonesia, using both a high-stakes field experiment that varied the extent of elite influence and nonexperimental data on a variety of existing government programs. While the relatives of those holding formal leadership positions are more likely to receive benefits in some programs, we argue that the welfare consequences of elite capture appear small: eliminating elite capture entirely would improve the welfare gains from these programs by less than one percent.
Keywords
proxy-means test (PMT)
experimental design
Authoring entity
| Agency Name | Role | Affiliation | |
|---|---|---|---|
| Vivi Alatas | Primary investigator | World Bank | valatas@worldbank.org |
| Abhijit Banerjee | Primary investigator | Department of Economics, MIT | banerjee@mit.edu |
| Rema Hanna | Primary investigator | Harvard Kennedy School | rema_hanna@hks.harvard.edu |
| Benjamin A. Olken | Primary investigator | Department of Economics, MIT | bolken@mit.edu |
| Ririn Purnamasari | Primary investigator | World Bank | rpurnamasari@worldbank.org |
| Matthew Wai-Poi | Primary investigator | World Bank | mwaipoi@worldbank.org |
Sponsor / Funding agency
| Name | Abbreviation | Grant number | Role |
|---|---|---|---|
| Australian Aid (World Bank Trust Fund) | AusAID | Financial support | |
| 3ie | OW3.1055 | Financial support | |
| NIH | P01 HD061315 | Financial support |
Other Identifications/ Acknowledgments
| Name | Role |
|---|---|
| Jurist Tan, Talitha Chairunissa, Amri Ilmma, Chaeruddin Kodir, He Yang, and Gabriel Zucker | Research assistance |
| Scott Guggenheim | Provided comments |
| Mitra Samya, BPS, TNP2K, and SurveyMeter | Field cooperation |
Topics
| ID | Topic | Vocabulary | Vocabulary URI |
|---|---|---|---|
| D72 | Political Processes: Rent-seeking, Lobbying, Elections, Legislatures, and Voting Behavior | JEL codes | https://www.aeaweb.org/econlit/jelCodes.php |
| H53 | National Government Expenditures and Welfare Programs | JEL codes | https://www.aeaweb.org/econlit/jelCodes.php |
| I38 | Welfare, Well-Being, and Poverty: Government Programs; Provision and Effects of Welfare Programs | JEL codes | https://www.aeaweb.org/econlit/jelCodes.php |
| O15 | Economic Development: Human Resources; Human Development; Income Distribution; Migration | JEL codes | https://www.aeaweb.org/econlit/jelCodes.php |
| O17 | Formal and Informal Sectors; Shadow Economy; Institutional Arrangements | JEL codes | https://www.aeaweb.org/econlit/jelCodes.php |
Language
English
Methods, software and scripts
Methods or algorithms applied
linear regression with large dummy-variable set (areg)
probit regression
Test linear hypotheses after estimation
Software
| Name | Version |
|---|---|
| Stata | 14 |
Scripts
|
File name
0MASTER-20190918.do
Zip archive
119802-V1.zip
Description
Master do file; this script calls all do files required to replicate the output from start to finish (in no more than a few minutes)
Authors
Format
Stata do file
software
Stata 14
Notes
Original data is in data-PUBLISH/originaldata and is all that is needed to run the code; all data in data-PUBLISH/codeddata is created from the coding do files. All results are then created and saved in output-PUBLISH/tables.
|
|||
|
File name
coding baseline.do
Zip archive
119802-V1.zip
Description
Coding/matching script 1/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
coding suseti pmt.do
Zip archive
119802-V1.zip
Description
Coding/matching script 2/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
coding elite relation.do
Zip archive
119802-V1.zip
Description
Coding/matching script 3/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
matching hybrid.do
Zip archive
119802-V1.zip
Description
Coding/matching script 4/7; Generates poverty density measure
Format
Stata do file
software
Stata 14
|
|||
|
File name
coding existing social programs.do
Zip archive
119802-V1.zip
Description
Coding/matching script 5/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
coding kitchen-sink variables.do
Zip archive
119802-V1.zip
Description
Coding/matching script 6/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
coding_partV_hybrid.do
Zip archive
119802-V1.zip
Description
Coding/matching script 7/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Table 1AB.do
Zip archive
119802-V1.zip
Description
Analysis script 1/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Table 1CD.do
Zip archive
119802-V1.zip
Description
Analysis script 2/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Table 2 Appendix Table 3.do
Zip archive
119802-V1.zip
Description
Analysis script 3/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Appendix Table 1A.do
Zip archive
119802-V1.zip
Description
Analysis script 4/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Appendix Table 1B.do
Zip archive
119802-V1.zip
Description
Analysis script 5/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Appendix Table 2.do
Zip archive
119802-V1.zip
Description
Analysis script 6/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
0 Appendix Table 4.do
Zip archive
119802-V1.zip
Description
Analysis script 7/7
Format
Stata do file
software
Stata 14
|
|||
|
File name
master_log_09182019.smcl
Zip archive
119802-V1.zip
Description
Latest log file obtained by running the master do file
Format
Stata log file
software
Stata 14
|
Reproduction instructions
The master do file should run start to finish in less than five minutes from the master do file '0MASTER 20190918.do'. Original data is in data-PUBLISH/originaldata and is all that is needed to run the code; all data in data-PUBLISH/codeddata is created from the coding do files. All results are then created and saved in output-PUBLISH/tables.
Key Subfolders:
1. code-PUBLISH: This folder contains all relevant code. The master do file is located here ('0Master20190918.do') as well as the two folders that are necessary for the creation of datasets/coding ('coding_matching' folder) and for the analysis/table creation ('analysis' folder). Users should update the directory on the master file to reflect the location of the directory on their computers once downloaded. Following that, all the data and output files needed to replicate the main findings of the paper (Tables 1A-1D, Table 2 and the 4 Appendix Tables) will be generated. The sub do files provide specific notes on the variables created where relevant.
2. data-PUBLISH: This folder contains all relevant .dta files. The first folder, 'original data' contains the 'Baseline' folder that has the original baseline survey information. Under 'original data' you will also find the 'Others' folder with the randomization results, the 2008 PPLS data and the PODES 2008 village level administrative data. The 'Endline2' folder contains the endline survey information. These datasets have been modified only to mask sensitive information. Finally, the 'codeddata' folder that stores intermediate datasets that are created through the sub 'coding_matching' do files.
3. log-PUBLISH: This folder contains the latest log file. When users run the master do file, a new log file will automatically be created and stored here.
4. output-PUBLISH: This folder contains all the tables of the main paper and appendix. When users run the master do file, these tables will be automatically overwritten.
Key Subfolders:
1. code-PUBLISH: This folder contains all relevant code. The master do file is located here ('0Master20190918.do') as well as the two folders that are necessary for the creation of datasets/coding ('coding_matching' folder) and for the analysis/table creation ('analysis' folder). Users should update the directory on the master file to reflect the location of the directory on their computers once downloaded. Following that, all the data and output files needed to replicate the main findings of the paper (Tables 1A-1D, Table 2 and the 4 Appendix Tables) will be generated. The sub do files provide specific notes on the variables created where relevant.
2. data-PUBLISH: This folder contains all relevant .dta files. The first folder, 'original data' contains the 'Baseline' folder that has the original baseline survey information. Under 'original data' you will also find the 'Others' folder with the randomization results, the 2008 PPLS data and the PODES 2008 village level administrative data. The 'Endline2' folder contains the endline survey information. These datasets have been modified only to mask sensitive information. Finally, the 'codeddata' folder that stores intermediate datasets that are created through the sub 'coding_matching' do files.
3. log-PUBLISH: This folder contains the latest log file. When users run the master do file, a new log file will automatically be created and stored here.
4. output-PUBLISH: This folder contains all the tables of the main paper and appendix. When users run the master do file, these tables will be automatically overwritten.
Datasets
Datasets
Coded data (intermediary data files generated by the scripts)
Data access type: Public
Stata 14 data files
Geographic locations
| Location | Code | Type |
|---|---|---|
| Indonesia | IDN | Country |
Disclaimer
You acknowledge that the original collector of the data, ICPSR, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
Confidentiality
The published materials do not contain confidential information.