Public willingness to pay for improvements in ecosystem services and landowner willingness to accept for wetlands conservation: An assessment of benefit transfer validity and reliability using choice experiments in several Canadian watersheds

Benefit-cost analyses are often used to evaluate the economic efficiency of proposed policies or projects. Such analyses require analysts to estimate the benefits and costs in monetary terms of any changes related to the policy being analyzed, including to the environment (e.g., changes in water or air quality). However, estimating these monetary values can be difficult since prices are often not available due to market failure. As such, several non-market valuation techniques have been developed for use in assessing these monetary values, including original research techniques, such as choice experiments, and benefit transfer which applies existing non-market values estimated using original research techniques to other contexts (e.g., locations). Several studies have evaluated the validity and reliability of benefit transfer in a variety of contexts. In this thesis, I contribute to this literature by assessing transfers in contexts not yet evaluated. In doing so, I use choice experiments to investigate landowner preferences for wetlands conservation in two Ontario watersheds and elicit the general public’s willingness to pay values for changes in ecosystem services in four Canadian watersheds. This research resulted in four papers. The first paper, motivated by the loss of wetlands in Southern Ontario, involves assessing the preferences and willingness to accept (WTA) of farm and non-farm landowners for enrolling their land in wetlands conservation programs. Though preferences and values are heterogeneous, many landowners are willing to enrol and at moderate cost. Using data from this paper, in the second and third papers I evaluate the validity and reliability of transfers of WTA and predicted program participation market shares, respectively. Results suggest that transfers of WTA are similarly valid and reliable to transfers of willingness to pay, while transfers of predicted participation market shares are considerably more valid and reliable than a parallel assessment of transfers of WTA. Finally, using data from the general public survey I evaluate alternatives for reconciling quantitative choice experiment attributes with differing levels for benefit transfer. A key finding of this research is that transfers rooted in “relative” preferences are more valid and reliable than transfers rooted in “absolute” preferences.