Incentive mechanisms and the provision of public goods: Field experiment data for testing alternative economic frameworks to supply ecosystem restoration on Virginia's Eastern Shore: 2008 data.

This dataset consists of participant responses in one of two economic experiments conducted on Virginia's Eastern Shore during 2008 and 2009 by Elizabeth C. Smith used to gauge resident preferences and willingness-to-pay for ecosystem restoration activities. This dataset was designed to be used to examine a practical method to implement an individualized pricing approach to public good provision, grounded in Lindahl's marginal benefit theory. The study's focus was on ecosystem valuation and market approaches that have potential to provide public goods, examining the potential to generate revenues for public goods from consumers. While willingness-to-pay measurement techniques have been used to assess preferences for many environmental goods, this research goes a step further to explore real money auctions that generate revenues sufficient to pay for restoration activities. The data from the field experiments conducted in coastal Virginia were used, along with laboratory experiment data, to evaluate the performance of auction mechanisms in generating revenues relative to potential (Hicksian) willingness to pay for marginal increments in public goods. The field execution of this experiment involved residents of Virginia's Eastern Shore and local public goods. This application involved half-acre increments of ecosystem restoration for sea grass habitat in coastal lagoons, plantings for migratory bird habitat, and, in some auctions, clam-based increments of water quality services, defined as delaying the harvest of clams for six months beyond normal harvest by an existing aquaculture firm. To perform these tasks, participants were provided a budget, between $90 and $150. The auctioneer described for participants the ecosystem services that may result from additional ecosystem restoration associated with each activity. The actual levels of ecosystem restoration provided were based on aggregate offers reaching a pre-determined (but unknown to the participants) provision point, or cost of implementing the project. The auction process considered offers to pay for and costs to deliver each unit in the sequence. In this way the auction process aggregated participants' willingness to pay (offers) on a given level of restoration and balanced aggregate marginal payment with marginal cost for any level of restoration provided. Data from the field experiment can be used to compare auction prices offered to estimates of marginal Hicksian willingness to pay derived from a baseline choice experiment that employed an incentive compatible, majority vote mechanism and actual (not hypothetical) money payments. A conditional logit model, rooted in McFadden's choice theory, can then be used to examine the trade-offs between various ecosystem restoration activities in order to estimate willingness to pay. The study for which this data was collected was intended to initiate development of new approaches for financing public goods, beyond government and philanthropic efforts. The study was also intended as a direct test of the long-held (90-year) assumption that individualized pricing based on the Lindahl approach is impractical in microeconomics.