Schooling and small stakes risk aversion: Insights from a rural-poor community

A risk experiment is conducted with 900 rural-poor subjects from Bangladesh to test the empirical basis for Rabin (2000)’s critique. By varying lab wealth within-subjects, this study provides additional evidence on whether subjects appear to be too risk-averse with regard to small-stakes gambles. I find risk aversion for low levels of lab wealth and insignificant distinction of risk aversion for high levels of lab wealth. Segregating the sample by schooling, while subjects with no schooling exhibited moderate levels of risk aversion throughout the experiment range, subjects with some formal schooling exhibited risk neutrality (or preference) for high levels of lab wealth. The latter result rejects the empirical basis for the critique. The findings stress the importance to better understand diverse risk attitudes to predict behavioral responses to interventions adequately.