Missing public funds and targeting performance: Evidence from an anti-poverty transfer program in Indonesia

This paper investigates how failing to consider missing funds in public resource delivery can lead to misleading conclusions on the nature and correlates of targeting performance. Combining administrative data on disbursement and household survey data on receipt under Indonesia's anti-poverty program, Inpres Desa Tertinggal (IDT), we find that only 69% of disbursements were actually received by the intended beneficiaries. When these missing funds are ignored, the distribution of IDT benefits is pro-poor, and better targeting is found in districts with higher per capita expenditure. However, when the missing funds are taken into account, the distribution of IDT benefits in fact becomes less pro-poor than universal, equal distribution, and better targeting is correlated with a higher proportion of female village heads and a higher level of villagers' human capital. These results underscore the importance of considering the size and allocation of missing funds in the analysis of targeting in public resource delivery.