Using survey-based choice experiments, we find that willingness to pay (WTP) for cap-and-trade programs depends upon their distributional impacts, including changes in the numbers of carbon-intensive versus green jobs and whether there will be additional regulations to limit non-global co-pollutant emissions from firms that buy permits. We estimate a model suitable for out-of-sample forecasting of WTP in other regions nationally, where systematic heterogeneity is captured by predicted county-level climate-change attitudes from the Yale Climate Map project.
A Heckman-type approach to sample selection correction is inappropriate for the conditional logit choice models used to analyze choice experiments. Logit model errors preclude reliance on the conventional assumption of potentially correlated bivariate normal errors. We propose and demonstrate a novel method of sample selection correction for multi-alternative conditional logit models that adapts mixed logit estimation methods.
Detection of systematic sample selection and strategies for correction
A model of recreation demand is developed to determine the role of water levels in determining participation at and frequency of trips taken to various federal reservoirs and rivers in the Columbia River Basin. Contingent behavior data are required to break the near-perfect multicollinearities among water levels at some waters.