Lab and life: Does risky choice behaviour observed in experiments reflect that in the real world?

Risk preferences play a crucial role in a great variety of economic decisions. Measuring risk preferences reliably is therefore an important challenge. In this paper we ask the question whether risk preferences observed in economic experiments reflect real-life risky choice behaviour. We investigate in a sample representative for a rural region of eastern Uganda whether pursuing farming strategies with both a higher expected profit and greater variance of profits is associated with willingness to take risks in an experiment. Controlling for other determinants of risk-taking in agriculture, we find that risky choice behaviour in the experiment is correlated with risky choice behaviour in real life in one domain, i.e. the purchase of fertiliser, but not in other domains, i.e. the growing of cash crops and market-orientation more broadly. Our findings suggest that economic experiments may be good at capturing real-world risky choice behaviour that is narrowly bracketed.