In environmental economics there is a growing literature on the valuation of environmental externalities using consumer willingness to pay, which is a relevant input for welfare analysis of projects targeting sustainability. Discrete choice models based on random utility maximization are particularly interesting for determining consumer valuation of environmental goods for which there is no market price. A large number of studies concerned with determining willingness-to-pay (WTP) measures using discrete choice models report only point estimates, without correct standard errors or other measures of uncertainty. However, the analysis of reliability of the estimates of interest - willingness to pay, consumer benefits, market shares, elasticities - is essential for inferring actual benefits.
In this paper, we contribute to the literature on characterizing the distribution of WTP measures by exploring Bayesian indifference on parameter ratios, and by analyzing the implications of considering individual random effects on the determination of confidence intervals. We show that these implications are not trivial and have an impact on how to summarize the WTP distributions. As an application, we study the distribution of the WTP for reducing CO2 emissions for two empirical situations: choice of heating versus insulation, and adoption of ultra-low emission vehicles. Therefore, this paper elucidates the value of Bayesian techniques for environmental evaluation.
Daziano, Ricardo A. and Martin Achtnicht (2013), Accounting for Uncertainty in Willingness to Pay for Environmental Benefits, ZEW Discussion Paper No. 13-059, Mannheim. Download