Indonesia’s Social Assistance Program Faces Scrutiny Amidst Mistargeting Concerns
Jakarta, Indonesia – A recent report indicates significant issues with the targeting of Indonesia’s flagship social assistance program, the Program Keluarga Harapan (PKH), raising questions about its effectiveness and sparking protests across the country. In September 2025, Indonesian media reported that as much as 45% of social assistance may have been distributed to ineligible recipients, representing an estimated 14.17 trillion rupiah in misallocated funds.
The PKH, established in 2005, is the second-largest Conditional Cash Transfer (CCT) program globally, currently reaching an estimated 40 million people – 10 million families. CCT programs, operating in 64 low- and middle-income countries, provide financial assistance to low-income families contingent upon meeting certain requirements, such as school attendance and regular health checkups. The intention is to invest in nutrition, health, and education while efficiently targeting those most in need.
While earlier assessments suggested the PKH had positive impacts, including reduced stunting, increased school enrollment, and improved maternal health outcomes, the recent findings have triggered a wave of criticism and prompted government action. Officials have pledged to digitize the system and strengthen oversight, initiating a verification process of 12 million recipients and removing those deemed ineligible. This mass removal, however, has led to protests and complaints from individuals who believe they were wrongly accused or victims of system errors.
The challenges faced by Indonesia’s PKH are not unique. Mexico, a pioneer in CCT programs, abolished its version after two decades of implementation, finding that poverty levels had not improved. The decision was influenced by a desire to reduce administrative burdens and address the impact of strict conditions placed on recipients.
A recent study undertaken by researchers, including the author, explored the consequences of CCT practices in rural Indonesia. The research combined interviews with officials, document analysis, and fieldwork within rural communities, utilizing community poverty ranking exercises, household surveys, and comparisons with econometric targeting methods. The study highlighted a “politics of knowledge” – the dominance of certain ideas about poverty that shape policy, often sidelining alternative understandings.
CCT programs rely on complex data systems and practices, including proxy means tests, social registries, and econometric algorithms, to identify and target impoverished households. However, the research revealed discrepancies between “objective” poverty data and the realities on the ground. Enumerators sometimes take shortcuts, village leaders struggle with pre-selected beneficiary lists, and respondents may strategically misrepresent their assets to appear poorer.
The system, researchers found, often simplifies complex social realities. Simple asset indicators fail to capture nuances of household welfare, and a fragmented, centralized system is prone to errors and data management problems. Coordination issues between government agencies can also lead to delays in fund disbursement.
In some communities studied, the PKH reached only a fraction of households identified as poor through separate village ranking exercises. Many recipients were found to be just above the official poverty line, while a significant number of food-insecure families received no benefits. This aligns with other studies suggesting exclusion rates of up to 60% for programs targeting the poorest 10% of the population, and that up to 52% of poor households may miss out on assistance.
The implementation of the PKH involves a process of translation, where program implementers and field officials adapt the program’s design to fit local contexts. This often leads to patterns of inclusion and exclusion that conflict with community perceptions of need and deservingness. Village leaders sometimes engage in informal redistribution to mitigate social tensions caused by exclusion, particularly when there’s a gap between technocratic targeting and local expectations of fairness.
While improving digital platforms and management information systems are often proposed as solutions to mistargeting, the core issue lies within the knowledge systems and measurement approaches used. Poverty programs inherently involve political decisions about who to help and how to address social inequality. By focusing on technical solutions, these programs can obscure the underlying political considerations and limit discussions about alternative approaches.
The CCT model addresses the symptoms of poverty by providing financial assistance, but it doesn’t fully address the underlying drivers of poverty in rural contexts. A more effective approach, researchers suggest, requires broadening the ways of knowing and measuring poverty, incorporating community understandings of need, and adopting more inclusive targeting methods. This could involve considering a more universal scheme or simple lifecycle transfers, rather than narrowly targeting the extremely poor. Effective social assistance requires aligning with local notions of fairness and social reciprocity.
Anti-poverty programs, the research concludes, fall short when they overlook the social, economic, and political realities that shape poverty in favor of purely technical solutions. Effective poverty reduction requires tackling the structural foundations of poverty while broadening our understanding of how poverty is experienced and measured.
