An important question to microfinance is the relevance of existing social capital in target communities to the performance of group lending. This research presents evidence from field experiments in South Africa and Armenia, in which subjects participate in trust games and a microfinance game. We present moderately strong evidence that personal trust between group members and peer homogeneity are more important to group loan repayment than general societal trust or mere acquaintanceship between members. We also find some evidence of reciprocity in groups: those who have been helped by other members are more likely to contribute themselves.