Robustness of inferences in risk and time experiments to lifecycle asset integration

Abstract

Experiment participants can engage in unobservable asset integration, mentally incorporating non-experimental “field” resources into an ostensibly controlled scenario. We extend asset integration to include lifecycle intertemporal tradeoffs. Our model shows that interference from purely exogenous field resources can be controlled with a simple wealth-level adjustment. Highly substitutable endogenous resources, by contrast, require modeling the entire experiment-field interaction. We examine the practical implications using three classic experiments on risk and time. As interference worsens, experimental decisions tend to exhibit an attenuation toward less risk aversion and more patience. This occurs reliably with household-scale resources, but pocket-change amounts can also cause problems.