Financial compensation and uncertainty: Using mean-variance rule and stochastic dominance to derive conservation payments for secondary forests

The expected opportunity costs of conserving a specific land use are usually considered adequate as financial compensation. However, a "conservation premium" is sometimes proposed as an added incentive, i.e., compensation greater than the expected opportunity costs. This paper discusses various methodological opportunities for deriving effective compensation under uncertainty. Based on cumulative distribution functions of possible opportunity costs (a Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) plantation was considered the alternative to conserving a Chilean secondary forest), generated through Monte Carlo simulations, we derived an inclusive range of possible compensations from 77 up to 375 US$·ha ^-1·year^-1. If we assumed that the two land-use alternatives were mutually exclusive and independent from other risky investments, a compensation of 375 US$·ha ^-1·year^-1 was necessary to convince every decision maker to maintain the secondary forest. However, only 77 US$· ha^-1·year^-1 was enough for a risk-averse decision maker (given average opportunity costs of 113 US$·ha ^-1·year^-1). Yet, it turned out that the greatest possible opportunity costs would already be compensated for with 199 US$·ha^-1·year^-1, given an error probability of 0.05. Compensating for the last 5% of possible opportunity costs would thus require an additional 176 US$·ha ^-1·year^-1. Our approach had two main limitations, namely we did not consider portfolio effects, which would allow diversifying away unsystematic risks, and we did not take into account the different systematic risks of the compared alternatives. These limitations may have led to an overestimation of effective compensation.