We propose an analytical distinction between standard risk aversion based on the valuation of a single gamble and marginal risk aversion based on the change in valuation between two gambles. We measure marginal risk aversion in two dimensions—mean and variance. Data from a field experiment is...

Estimating risk preferences is tricky because controlling for confounding factors is difficult. Omitting or imperfectly controlling for these factors can attribute too much observable behaviour to risk aversion and bias estimated preferences. Agents often modify risky decisions in response to...