This paper explores whether results from experimental economics games can predict real-world financial decisions and whether social capital can help solve credit market failures. Dean S. Karlan conducted two experiments, a Trust game and a Public Goods game, along with a survey to measure social capital. He then examined whether behavior in these games predicted repayment of loans in a Peruvian microfinance program. The results showed that individuals identified as "trustworthy" in the Trust game were less likely to default on their loans. However, individuals identified as more "trusting" saved less and had higher repayment problems. The paper also found that answers to the General Social Survey (GSS) questions on trust, fairness, and helping others correlated with real financial outcomes. Individuals who answered the GSS questions more positively were more likely to repay their loans and save more. The Public Goods game results showed that individuals who contributed more were not necessarily more likely to repay their loans. The paper concludes that the Trust game can predict repayment behavior, but not necessarily trust. It also highlights the importance of social capital in group lending programs and the role of trustworthiness in financial decisions. The study provides insights into the determinants of default and savings for participants in a group banking project for the poor.This paper explores whether results from experimental economics games can predict real-world financial decisions and whether social capital can help solve credit market failures. Dean S. Karlan conducted two experiments, a Trust game and a Public Goods game, along with a survey to measure social capital. He then examined whether behavior in these games predicted repayment of loans in a Peruvian microfinance program. The results showed that individuals identified as "trustworthy" in the Trust game were less likely to default on their loans. However, individuals identified as more "trusting" saved less and had higher repayment problems. The paper also found that answers to the General Social Survey (GSS) questions on trust, fairness, and helping others correlated with real financial outcomes. Individuals who answered the GSS questions more positively were more likely to repay their loans and save more. The Public Goods game results showed that individuals who contributed more were not necessarily more likely to repay their loans. The paper concludes that the Trust game can predict repayment behavior, but not necessarily trust. It also highlights the importance of social capital in group lending programs and the role of trustworthiness in financial decisions. The study provides insights into the determinants of default and savings for participants in a group banking project for the poor.