This paper investigates the relationships between inequality, elite capture, and targeting performance of two biggest social protection programs in Indonesia, the rice for the poor (Raskin) and direct cash transfer (BLT) programs. Both programs differ in their targeting methods. While targeting in Raskin is decentralized, targeting in BLT is more centralized. Using data from 2009 National Socioeconomic Household Survey (Susenas) and 2008 Village Census (Podes), we find that an increase in Gini ratio is not significantly associated with a change in inclusion error in both programs, indicating the existence of elite capture in both programs. However, an increase of 0.01 point in Gini ratio is associated with a reduction of 0.55 percentage point in exclusion error of BLT, while the elasticity is smaller in Raskin with only 0.50, implying a bigger elite capture in this program.