Local Elites and Targeted Social Programs in Indonesia
What is the extent of elite capture in targeted social programs and does it justify excluding local leaders from program implementation? Researchers conducted a randomized evaluation to study the impact of elite capture in a conditional cash transfer program in Indonesia. In addition, they analyzed historical administrative data to evaluate other targeted social programs. Elite capture existed in some programs, but the welfare losses created were small, suggesting that improving administrative capabilities in targeting methods might be more effective in enhancing social assistance programs than focusing on elite capture.
Local leaders often have better information than central governments about which members of their communities should receive benefits from targeted social programs. However, governments and donors are often reluctant to devolve decision making about choosing beneficiaries to local leaders, preferring to allocate benefits based on more objective, but sometimes less precise, measures. A key reason for this reluctance is concern over “elite capture,” in which local leaders take resources intended for other beneficiaries. As a result, many programs exclude local leaders from targeting decisions and do not incorporate their local knowledge. What is the extent of elite capture in targeted social programs and does it justify excluding local leaders from program implementation?
Context of the evaluation
The Government of Indonesian runs a number of household-targeted social assistance programs, with a total annual cost of over US$2.7 billion in 2009. Some of the largest targeted social programs include Raskin, which provides the poor with subsidized rice; Jamkesmas, which provides health insurance for the poor; and Bantuan Langsung Tunai (BLT), which provided monthly unconditional cash transfers for one year in 2005 and for nine months in 2008. In 2007, the Indonesian government introduced a conditional cash transfer program, Program Keluarga Harapan (PKH), which provides direct cash to beneficiaries ranging from Rp. 600,000 to 2.2 million (US$67-250) per year for six years, conditional on family composition, school attendance, pre and postnatal check-ups, and completed vaccinations.
Targeting for all four of these programs typically combines a data-based component with local input, where village elites play an important role.
Details of the intervention
Researchers conducted a randomized evaluation to study the impact of elite capture in Program Keluarga Harapan (PKH). In addition, they analyzed historical administrative data from the Government of Indonesia to test for elite capture within Raskin, Jamkesmas, and BLT.
Working in collaboration with the Government of Indonesia, researchers randomly varied the rules for beneficiary selection for the 2011 expansion of PKH for 400 randomly selected villages in Lampung, South Sumatra, and Central Java provinces.
Comparison villages: Beneficiaries were selected through the current status quo for PKH, with beneficiaries verified by surveys conducted by the central government on assets and other household characteristics (200 villages).
Elite-input villages: Beneficiaries were selected through a meeting of 5-8 local leaders, both formal (those who hold formal leadership positions such as village heads) and informal (respected members of the community), with no additional verification (100 villages).
Community-input villages: Beneficiaries were selected through a community meeting with no additional verification. The full community was invited to the meetings, which were heavily advertised (100 villages).
Results and policy lessons
Researchers found little evidence of elite capture. Where it existed, it was concentrated among formal elites and did not have large effects on overall welfare. These results indicate that focusing on limiting elite capture may be less important than improving administrative practices for enhancing overall service delivery.
Presence of elite capture: There was no evidence of elite capture in the PKH program. Local leaders and their relatives were not more likely to receive aid in all treatment groups, even when local leaders decided on beneficiary lists behind closed doors. Similarly, there was no evidence of elite capture within Raskin, Jamkesmas, and BLT, which researchers analyzed with historical administrative data.
Formal vs. informal elites: There was some evidence of elite capture among formal elites for the BLT, Jamkesmas, and Raskin programs, but not in PKH. Formal leaders and their relatives were 5 percentage points (14 percent) more likely to receive BLT in both 2005 and 2008 compared to 36 percent of all households; 8 percentage points (19 percent) more likely to receive Jamkesmas compared to 43 percent of all households; and 3 percentage points (4 percent) more likely to receive Raskin compared to 75 percent of all households. This elite capture occurred during the implementation of the program, rather than during the official targeting process. In contrast, informal leaders and their relatives were less likely to receive targeted benefits than expected across almost all the programs considered.
Types of programs subject to elite capture: Capture by formal elites occurred in programs that were targeted at roughly 40 percent of the population but did not occur in PKH, which targets the bottom 5 percent. Furthermore, elite capture was most likely to occur in villages that received more slots for programs relative to the number of poor residents in the village. This suggests that elite capture is more likely to occur when there are additional benefits left over after addressing the very poor or when relatively well-off households can plausibly claim to be eligible for the program.
Impact on welfare: While elite capture existed in some programs, the quantitative and economic impact on overall welfare was small. Formal elites comprised a small share of the population (at most 15 percent), were not much richer than non-elites (about 9 percent richer), and were at most about 8 percentage points more likely to receive benefits. Researchers estimate that eliminating elite capture entirely would improve the welfare gains from these programs by less than one percent. In contrast, researchers estimate that improving administrative capabilities in targeting methods could improve program welfare gains by as much as 26 to 31 percent. These results indicate that eliminating elite capture would not improve welfare as much as improving administrative capabilities.
Alatas, Vivi, Abhijit Banerjee, Rema Hanna, Ben Olken, Ririn Purnamasari, and Matthew Wai-Poi. “Does Elite Capture Matter? Local Elites and Targeted Welfare Programs in Indonesia.” NBER Working Paper #18798, Cambridge, February 2013.