This paper examines China's ruralminimum living standard guarantee (dibao) program, one ofthe largest targeted transfer schemes in the world. Usinghousehold survey data matched with published administrativedata, the authors provide background on the patterns ofinequality and poverty in rural China, describe the dibaoprogram, estimate the program's impact on poverty, andcarry out targeting analysis. The authors find that theprogram provides sufficient income to poor beneficiaries butdoes not substantially reduce the overall level of poverty,in part because the number of beneficiaries is smallrelative to the number of poor. Conventional targetinganalysis reveals rather large inclusionary and exclusionarytargeting errors; propensity score targeting analysis yieldssmaller but still large targeting errors. Simulations ofpossible reforms to the dibao program indicate thatexpanding coverage can potentially yield greater povertyreduction than increasing transfer amounts. In addition,replacing locally diverse dibao lines with a nationallyuniform dibao threshold can in theory reduce poverty. Thepotential gains in poverty reduction, however, depend on theeffectiveness of targeting.