Questions remain as to whether results from experimental economics games are generalizable to real decisions in non-laboratory settings. Furthermore, important questions persist about whether social capital can help solve seemingly missing credit markets. I conduct two experiments, a Trust game and a Public Goods game, and a survey to measure social capital. I then examine whether behavior in the games predicts repayment of loans to a Peruvian group lending microfinance program. Since the structure of these loans relies heavily on social capital to enforce repayment, this is a relevant and important test of the games, as well as of other measures of social capital. I find that individuals identified as "trustworthy" by the Trust game are in fact less likely to default on their loans. I do not find similar support for the Trust game as a measure of trust.