Abstract
Environmental policy evaluation is often criticised for employing discount rates that have little grounding in research. Yet, experimental research aimed at eliciting realistic rates will inevitably require strong assumptions of external validity, while also placing large cognitive demands on subjects by processing tasks of increased unfamiliarity. We use a controlled lab experiment to test the impact of incentives on risk aversion and discounting tasks for monetary and environmental goods. We find that, on average, incentives have little effect on risk aversion or discounting tasks in either domain. Exploring heterogeneity by treatment and socio-demographics some significant patterns emerge. Further, contrary to past work, we find evidence of domain (monetary vs. environmental good) effects in both risk and discounting.
Highlights
Choices over uncertain outcomes that play out over time permeate most individual and societal decisions
We find evidence of domain effects in both risk and discounting, with subjects exhibiting higher risk aversion and higher discount rates in the environmental domain
We review past work in an attempt to bring together findings from the experimental literature on payoff effects as well as findings from the environmental economics literature on hypothetical bias given the crossover in the relevance of these two strands of literature
Summary
Choices over uncertain outcomes that play out over time permeate most individual and societal decisions. With future outcomes embedding an element of uncertainty they are further penalised as individuals recognize their probabilistic nature These characteristics, impatience and risk aversion, are mirrored in environmental and health policy appraisals where discount rates, encompassing time preference and risk aversion, are applied to benefits occurring in the future. Since risk aversion tasks measure preferences rather than performance, there is no way of assessing whether incentives improve performance other than to monitor differences between the two treatments or any deviation from a priori expectations of economic theory. A review of 74 experiments comparing different levels of financial incentives found no effect of financial incentives on mean performance but did note a reduction in variance and a reduction in presentation effects (Camerer & Hogarth, 1999)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
