Private transfer payments are modelled as outcomes of a constrained social choice problem facing donors. The approach is applied to a large household level data set for Java and hypotheses are tested concerning the performance of the "moral economy" as a social security system. Transfer behavior is found to be very different between rural and urban areas. While transfer receipts and outlays are income inequality reducing in rural areas, this is not the case in urban areas. There is also evidence of transfers being targeted to disadvantaged households such as the sick, elderly and (for urban areas) the unemployed.