Consistencia en la elección de políticas ambientales con efectos en la salud

Resumo

This article analyses the impact of inconsistent behaviour on the results of a discrete choice experiment used to evaluate policies aiming to reduce air pollution-related health hazards created by a steam power plant. This involved differentiating two types of inconsistency, depending on whether respondents failed the dominance or the stability test built into the successive choices. Random parameter and heteroscedastic models were used for the analysis. The results show that the presence of inconsistencies affects the magnitude and accuracy of welfare measures, increasing the willingness to pay and lowering efficiency. Changes were also recorded in the relative importance of the various attributes. Tests to detect inconsistencies must be included in discrete choice experiments and certain aspects of their design must be formulated very carefully to reduce complexity and consequently the likelihood of inconsistencies.